draft-ietf-ccamp-gmpls-recovery-analysis-00.txt   draft-ietf-ccamp-gmpls-recovery-analysis-01.txt 
CCAMP Working Group CCAMP GMPLS P&R Design Team CCAMP Working Group CCAMP GMPLS P&R Design Team
Internet Draft Internet Draft
Expiration Date: June 2003 Dimitri Papadimitriou (Editor) Category: Informational Dimitri Papadimitriou (Editor)
Eric Mannie (Editor) Expiration Date: November 2003 Eric Mannie (Editor)
January 2003 May 2003
Analysis of Generalized MPLS-based Recovery Mechanisms Analysis of Generalized MPLS-based Recovery Mechanisms
(including Protection and Restoration) (including Protection and Restoration)
draft-ietf-ccamp-gmpls-recovery-analysis-00.txt draft-ietf-ccamp-gmpls-recovery-analysis-01.txt
Status of this Memo Status of this Memo
This document is an Internet-Draft and is in full conformance with This document is an Internet-Draft and is in full conformance with
all provisions of Section 10 of RFC2026 [1]. all provisions of Section 10 of RFC2026 [1].
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet- other groups may also distribute working documents as Internet-
Drafts. Internet-Drafts are draft documents valid for a maximum of Drafts. Internet-Drafts are draft documents valid for a maximum of
skipping to change at line 43 skipping to change at line 43
For potential updates to the above required-text see: For potential updates to the above required-text see:
http://www.ietf.org/ietf/1id-guidelines.txt http://www.ietf.org/ietf/1id-guidelines.txt
1. Abstract 1. Abstract
This document provides an analysis grid that can be used to This document provides an analysis grid that can be used to
evaluate, compare and contrast the numerous Generalized MPLS evaluate, compare and contrast the numerous Generalized MPLS
(GMPLS)-based recovery mechanisms currently proposed at the CCAMP (GMPLS)-based recovery mechanisms currently proposed at the CCAMP
Working Group. A detailed analysis of each of the recovery phases is Working Group. A detailed analysis of each of the recovery phases is
provided using the terminology defined in [CCAMP-TERM]. Also, this provided using the terminology defined in a companion document. This
document focuses on transport plane survivability and recovery document focuses on transport plane survivability and recovery
issues and not on control plane resilience and related aspects. issues and not on control plane resilience and related aspects.
D.Papadimitriou et al. - Internet Draft ű Expires June 2003 1 D.Papadimitriou et al. - Internet Draft - Expires November 2003 1
2. Contributors 2. Contributors
This document is the result of the CCAMP Working Group Protection This document is the result of the CCAMP Working Group Protection
and Restoration design team joint effort. Besides the editors, the and Restoration design team joint effort. Besides the editors, the
following are the authors that contributed to the present memo: following are the authors that contributed to the present memo:
Deborah Brungard (AT&T) Deborah Brungard (AT&T)
Rm. D1-3C22 - 200 S. Laurel Ave. Rm. D1-3C22 - 200 S. Laurel Ave.
Middletown, NJ 07748, USA Middletown, NJ 07748, USA
E-mail: dbrungard@att.com E-mail: dbrungard@att.com
Sudheer Dharanikota (Consult) Sudheer Dharanikota (Consult)
E-mail: sudheer@ieee.org E-mail: sudheer@ieee.org
Jonathan P. Lang (Calient) Jonathan P. Lang (Consult)
25 Castilian E-mail: jplang@ieee.org
Goleta, CA 93117, USA
E-mail: jplang@calient.net
Guangzhi Li (AT&T) Guangzhi Li (AT&T)
180 Park Avenue, 180 Park Avenue,
Florham Park, NJ 07932, USA Florham Park, NJ 07932, USA
E-mail: gli@research.att.com E-mail: gli@research.att.com
Eric Mannie (Consult)
E-mail: eric_mannie@hotmail.com
Dimitri Papadimitriou (Alcatel)
Francis Wellesplein, 1
B-2018 Antwerpen, Belgium
E-mail: dimitri.papadimitriou@alcatel.be
Bala Rajagopalan (Tellium) Bala Rajagopalan (Tellium)
2 Crescent Place - P.O. Box 901 2 Crescent Place - P.O. Box 901
Oceanport, NJ 07757-0901, USA Oceanport, NJ 07757-0901, USA
E-mail: braja@tellium.com E-mail: braja@tellium.com
Yakov Rekhter (Juniper) Yakov Rekhter (Juniper)
1194 N. Mathilda Avenue
Sunnyvale, CA 94089, USA
E-mail: yakov@juniper.net E-mail: yakov@juniper.net
Conventions used in this document:
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in
this document are to be interpreted as described in RFC-2119 [2].
Any other recovery-related terminology used in this document
conforms to the one defined in [CCAMP-TERM]. The reader is also
assumed to be familiar with the terminology developed in [GMPLS-
ARCH], [RFC-3471], [GMPLS-RTG] and [LMP].
D.Papadimitriou et al. - Internet Draft - Expires November 2003 2
3. Introduction 3. Introduction
This document provides an analysis grid to evaluate, compare and This document provides an analysis grid to evaluate, compare and
contrast the numerous Generalized MPLS (GMPLS) based recovery contrast the numerous Generalized MPLS (GMPLS) based recovery
mechanisms currently proposed in the CCAMP Working Group. Here, the mechanisms currently proposed in the CCAMP Working Group. Here, the
focus will only be on transport plane survivability and recovery focus will only be on transport plane survivability and recovery
issues and not on control plane resilience related aspects. Although issues and not on control plane resilience related aspects. Although
the recovery mechanisms described in this document impose different the recovery mechanisms described in this document impose different
requirements on recovery protocols, the protocol(s) specifications requirements on GMPLS-based recovery protocols, the protocol(s)
will not be covered in this document. Though the concepts discussed specifications will not be covered in this document. Though the
here are technology independent, this document will implicitly focus concepts discussed here are technology independent, this document
on SONET/SDH and pre-OTN technologies except when specific details will implicitly focus on Sonet/SDH and pre-OTN technologies except
need to be considered (for instance, in the case of failure when specific details need to be considered (for instance, in the
detection). Details for applicability to other technologies such as case of failure detection). Details for applicability to other
Optical Transport Networks (OTN) [ITUT-G709] will be covered in a technologies such as Optical Transport Networks (OTN) [G.709] will
future release of this document. be covered in a future release of this document.
In the present release, a detailed analysis is provided for each of In the present release, a detailed analysis is provided for each of
the recovery phases as identified in [CCAMP-TERM]. These phases the recovery phases as identified in [CCAMP-TERM]. These phases
define the sequence of generic operations that need to be performed define the sequence of generic operations that need to be performed
D.Papadimitriou et al. - Internet Draft ű June 2003 2
when a LSP/Span failure (or any other event generating such when a LSP/Span failure (or any other event generating such
failures) occurs: failures) occurs:
- Phase 1: Failure detection - Phase 1: Failure detection
- Phase 2: Failure localization and isolation - Phase 2: Failure localization and isolation
- Phase 3: Failure notification - Phase 3: Failure notification
- Phase 4: Recovery (Protection/Restoration) - Phase 4: Recovery (Protection/Restoration)
- Phase 5: Reversion (normalization) - Phase 5: Reversion (normalization)
Failure detection, localization and notification phases together are Failure detection, localization and notification phases together are
referred to as fault management. Within a recovery domain, the referred to as fault management. Within a recovery domain, the
entities involved during the recovery operations are defined in entities involved during the recovery operations are defined in
[CCAMP-TERM]; these entities include ingress, egress and [CCAMP-TERM]; these entities include ingress, egress and
intermediate nodes. intermediate nodes.
In this document the term ˘recovery mechanism÷ is used to cover both In this document, the term "recovery mechanism" is used to cover
protection and restoration mechanisms. Specific terms such as both protection and restoration mechanisms. Specific terms such as
protection and restoration are only used when differentiation is protection and restoration are only used when differentiation is
required. Likewise the term ˘failure÷ is used to represent both required. Likewise the term "failure" is used to represent both
signal failure and signal degradation. In addition, a clear signal failure and signal degradation. In addition, a clear
distinction is made between partitioning (horizontal hierarchy) and distinction is made between partitioning (horizontal hierarchy) and
layering (vertical hierarchy) when analyzing hierarchical recovery layering (vertical hierarchy) when analyzing hierarchical recovery
mechanisms including disjointness related issues. We also introduce mechanisms including disjointness related issues. We also introduce
the dimensions from which each of the recovery mechanisms described the dimensions from which each of the recovery mechanisms described
in this document can be further analyzed and provide an analysis in this document can be further analyzed and provide an analysis
grid with respect to these dimensions. Last, we conclude by grid with respect to these dimensions. Last, we conclude by
detailing the applicability of the current GMPLS protocol building detailing the applicability of the current GMPLS protocol building
blocks for recovery purposes. blocks for recovery purposes.
Any other recovery-related terminology used in this document
conforms to the one defined in [CCAMP-TERM].
Conventions used in this document:
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in
this document are to be interpreted as described in RFC-2119 [2].
4. Fault Management 4. Fault Management
4.1 Failure Detection 4.1 Failure Detection
D.Papadimitriou et al. - Internet Draft - Expires November 2003 3
Transport failure detection is the only phase that can not be Transport failure detection is the only phase that can not be
achieved by the control plane alone since the latter needs a hook to achieved by the control plane alone since the latter needs a hook to
the transport plane to collect the related information. It has to be the transport plane to collect the related information. It has to be
emphasized that even if failure events themselves are detected by emphasized that even if failure events themselves are detected by
the transport plane, the latter, upon failure condition, MUST the transport plane, the latter, upon failure condition, MUST
trigger the control plane for subsequent actions through the use of trigger the control plane for subsequent actions through the use of
GMPLS signalling capabilities (see [GMPLS-SIG]) or Link Management GMPLS signalling capabilities (see [RFC-3471]) or Link Management
Protocol (see [LMP], Section 6) capabilities. Protocol capabilities (see [LMP], Section 6).
Therefore, by definition, transport failure detection is transport Therefore, by definition, transport failure detection is transport
technology dependent (and so exceptionally, we keep here the technology dependent (and so exceptionally, we keep here the
"transport plane" terminology). In transport fault management,
D.Papadimitriou et al. - Internet Draft ű June 2003 3
˘transport plane÷ terminology). In transport fault management,
distinction is made between a defect and a failure. Here, the distinction is made between a defect and a failure. Here, the
discussion addresses failure detection (persistent fault cause). In discussion addresses failure detection (persistent fault cause). In
the technology dependent descriptions, a more precise specification the technology dependent descriptions, a more precise specification
will be provided. will be provided.
As an example, SONET/SDH (see [G.707], [G.783] and [G.806]) provides As an example, Sonet/SDH (see [G.707], [G.783] and [G.806]) provides
supervision capabilities covering: supervision capabilities covering:
- Continuity: monitors the integrity of the continuity of a trail - Continuity: monitors the integrity of the continuity of a trail
(i.e. section or path). This operation is performed by monitoring (i.e. section or path). This operation is performed by monitoring
the presence/absence of the signal. Examples are Loss of Signal the presence/absence of the signal. Examples are Loss of Signal
(LOS) detection for the physical layer, Unequipped (UNEQ) Signal (LOS) detection for the physical layer, Unequipped (UNEQ) Signal
detection for the path layer, Server Signal Fail Detection (e.g. detection for the path layer, Server Signal Fail Detection (e.g.
AIS) at the client layer. AIS) at the client layer.
- Connectivity: monitors the integrity of the routing of the signal - Connectivity: monitors the integrity of the routing of the signal
skipping to change at line 200 skipping to change at line 209
- Payload type: checks that compatible adaptation functions are used - Payload type: checks that compatible adaptation functions are used
at the source and the sink. This is normally done by adding a at the source and the sink. This is normally done by adding a
signal type identifier at the source adaptation function and signal type identifier at the source adaptation function and
comparing it with the expected identifier at the sink. For comparing it with the expected identifier at the sink. For
instance, the payload signal label and the corresponding payload instance, the payload signal label and the corresponding payload
signal mismatch detection. signal mismatch detection.
- Signal Quality: monitors the performance of a signal. For - Signal Quality: monitors the performance of a signal. For
instance, if the performance falls below a certain threshold a instance, if the performance falls below a certain threshold a
D.Papadimitriou et al. - Internet Draft - Expires November 2003 4
defect ű excessive errors (EXC) or degraded signal (DEG) - is defect ű excessive errors (EXC) or degraded signal (DEG) - is
detected. detected.
The most important point to keep in mind is that the supervision The most important point is that the supervision processes and the
processes and the corresponding failure detection (used to initiate corresponding failure detection (used to initiate the recovery
the recovery phase(s)) result in either: phase(s)) result in either:
- Signal Degrade (SD): A signal indicating that the associated data - Signal Degrade (SD): A signal indicating that the associated data
has degraded in the sense that a degraded defect condition is has degraded in the sense that a degraded defect condition is
active (for instance, a dDEG declared when the Bit Error Rate active (for instance, a dDEG declared when the Bit Error Rate
exceeds a preset threshold). exceeds a preset threshold).
D.Papadimitriou et al. - Internet Draft ű June 2003 4
- Signal Fail (SF): A signal indicating that the associated data has - Signal Fail (SF): A signal indicating that the associated data has
failed in the sense that a signal interrupting near-end defect failed in the sense that a signal interrupting near-end defect
condition is active (as opposed to the degraded defect). condition is active (as opposed to the degraded defect).
In Optical Transport Networks (OTN) equivalent supervision In Optical Transport Networks (OTN) equivalent supervision
capabilities are provided at the optical/digital section layers capabilities are provided at the optical/digital section layers
(OTS, OMS and OTUk) and at optical/digital path layers (OCh and (OTS, OMS and OTUk) and at optical/digital path layers (OCh and
ODUk). Interested readers are referred to the ITU-T Recommendations ODUk). Interested readers are referred to the ITU-T Recommendations
[G.798] and [G.709] for more details. [G.798] and [G.709] for more details.
The above are examples illustrate cases where the failure detection, The above are examples that illustrate cases where the failure
reporting and recovery responsible entities are co-located. detection, and reporting entities are co-located. The following
example illustrates the scenario where the failure detection and
On the other hand, in pre-OTN networks, a failure may be masked by reporting entities are not co-located.
intermediate O/E/O based Optical Line System (OLS), preventing a
Photonic Cross-Connect (PXC) from detecting upstream failures. In
such cases, failure detection may be assisted by an out-of-band
communication channel and failure condition reported to the PXC
control plane. This can be provided by using [LMP-WDM] extensions
that delivers IP message-based communication between the PXC and the
OLS control plane. Also, since PXCs are framing format independent,
failure conditions can only be triggered either by detecting the
absence of the optical signal or by measuring its optical quality,
mechanisms which are less reliable than electrical (digital)
mechanisms. Both types of detection mechanisms are out of the scope
of this document. If the intermediate OLS supports electrical
(digital) mechanisms, using the LMP communication channel, these
failure conditions are reported to the PXC and subsequent recovery
actions performed as described in Section 5. As such from the
control plane viewpoint, this mechanism makes the OLS-PXC composed
system appearing as a single logical entity allowing considering for
such entity the same failure management mechanisms as for any other
O/E/O capable device.
This example is to illustrate the scenario where the failure In pre-OTN networks, a failure may be masked by intermediate O/E/O
detection, reporting and recovery responsible entities are not co- based Optical Line System (OLS), preventing a Photonic Cross-Connect
located. (PXC) from detecting upstream failures. In such cases, failure
detection may be assisted by an out-of-band communication channel
and failure condition reported to the PXC control plane. This can be
provided by using [LMP-WDM] extensions that delivers IP message-
based communication between the PXC and the OLS control plane. Also,
since PXCs are framing format independent, failure conditions can
only be triggered either by detecting the absence of the optical
signal or by measuring its quality. These mechanisms are generally
less reliable than electrical (digital) ones. Both types of
detection mechanisms are out of the scope of this document. If the
intermediate OLS supports electrical (digital) mechanisms, using the
LMP communication channel, these failure conditions are reported to
the PXC and subsequent recovery actions performed as described in
Section 5. As such from the control plane viewpoint, this mechanism
makes the OLS-PXC composed system appearing as a single logical
entity allowing considering for such entity the same failure
management mechanisms as for any other O/E/O capable device.
More generally, the following are typical failure conditions in More generally, the following are typical failure conditions in
Sonet/SDH and pre-OTN networks: Sonet/SDH and pre-OTN networks:
- Loss of Light (LOL)/Loss of Signal (LOS): Signal Failure (SF) - Loss of Light (LOL)/Loss of Signal (LOS): Signal Failure (SF)
condition where the optical signal is not detected anymore on a condition where the optical signal is not detected anymore on a
given interfaceĂs receiver. given interface's receiver.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 5
- Signal Degrade (SD): detection of the signal degradation over - Signal Degrade (SD): detection of the signal degradation over
a specific period of time. a specific period of time.
- For Sonet/SDH payloads, all of the above-mentioned supervision - For Sonet/SDH payloads, all of the above-mentioned supervision
capabilities can be used, resulting in SD or SF condition. capabilities can be used, resulting in SD or SF condition.
In summary, the following cases are considered to illustrate the In summary, the following cases apply when considering the
communication between the detecting and reporting (also recovery communication between the detecting and reporting entities:
responsible) entities:
D.Papadimitriou et al. - Internet Draft ű June 2003 5
- Co-located detecting and reporting entities: both the detecting - Co-located detecting and reporting entities: both the detecting
and reporting entities are on the same node (e.g., Sonet/SDH and reporting entities are on the same node (e.g., Sonet/SDH
equipment, Opaque cross-connects, and, with some limitations, for equipment, Opaque cross-connects, and, with some limitations,
Transparent cross-connects, etc.). Transparent cross-connects, etc.)
- Non co-located detecting and reporting entities: - Non co-located detecting and reporting entities:
- with In-band communication between entities: - with in-band communication between entities: entities are
Entities are separated but transport plane (in-band) physically separated but the transport plane provides in-band
communication is provided between them (e.g., Server Signal communication between them (e.g., Server Signal Failures (AIS),
Failures (AIS), etc.) etc.)
- with Out-of-band communication between entities: - with out-of-band communication between entities: entities are
Entities are separated but out-of-band communication is provided physically separated but an out-of-band communication channel is
between them (e.g., using [LMP]). provided between them (e.g., using [LMP]).
4.2 Failure Localization and Isolation 4.2 Failure Localization and Isolation
Failure localization provides the required information in order to Failure localization provides to the deciding entity information
perform the subsequent recovery action(s) at the LSP/span end- about the location (and so the identity) of the transport plane
points. entity that detects the LSP(s)/span(s) failure. The deciding entity
can then take accurate decision to achieve finer grained recovery
switching action(s). Note that this information can also be included
as part of the failure notification (see Section 4.3).
In some cases, accurate failure localization may be less urgent; the In some cases, this accurate failure localization information may be
need is to identify the failure as occurring within the recovery less urgent to determine if it requires performing more time
domain. This is particularly the case when edge-to-edge LSP recovery consuming failure isolation (see also Section 4.5). This is
(edge referring to a sub-network end-node for instance) is performed particularly the case when edge-to-edge LSP recovery (edge referring
based on a simple failure notification (including the identification to a sub-network end-node for instance) is performed based on a
of the failed working LSPs) so that a more accurate localization can simple failure notification (including the identification of the
be performed after LSP recovery. working LSPs under failure condition). In this case, a more accurate
localization and isolation can be performed after recovery of these
LSPs.
Failure localization should be triggered immediately after the fault Failure localization should be triggered immediately after the fault
detection phase. This operation can be performed at the transport detection phase. This operation can be performed at the transport
management plane and/or, if unavailable (via the transport plane), plane and/or, if unavailable (via the transport plane), the control
the control plane level where dedicated signaling messages can be plane level where dedicated signaling messages can be used. When
used. performed at the control plane level, a protocol such as LMP (see
[LMP], Section 6) can be used for failure localization purposes.
When performed at the control plane level, a protocol such as LMP
(see [LMP], Section 6) can be used for failure localization and
isolation purposes.
4.3 Failure Notification 4.3 Failure Notification
Failure notification is used 1) to inform intermediate nodes that a Failure notification is used 1) to inform intermediate nodes that a
LSP/span failure has occurred and has been detected 2) to inform the LSP/span failure has occurred and has been detected 2) to inform the
recovery deciding entities (which can correspond to any intermediate recovery deciding entities (which can correspond to any intermediate
D.Papadimitriou et al. - Internet Draft - Expires November 2003 6
or end-point of the failed LSP/span) that the corresponding service or end-point of the failed LSP/span) that the corresponding service
is not available. In general, these deciding entities will be the is not available. In general, these deciding entities will be the
ones taking the appropriate recovery decision. When co-located with ones taking the appropriate recovery decision. When co-located with
the recovering entity, these entities will also perform the the recovering entity, these entities will also perform the
corresponding recovery action(s). corresponding recovery action(s).
Failure notification can be either provided by the transport or by Failure notification can be either provided by the transport or by
the control plane. As an example, let us first briefly describe the the control plane. As an example, let us first briefly describe the
failure notification mechanism defined at the SDH/SONET transport failure notification mechanism defined at the Sonet/SDH transport
plane level (also referred to as maintenance signal supervision): plane level (also referred to as maintenance signal supervision):
D.Papadimitriou et al. - Internet Draft ű June 2003 6
- AIS (Alarm Indication Signal) occurs as a result of a failure - AIS (Alarm Indication Signal) occurs as a result of a failure
condition such as Loss of Signal and is used to notify downstream condition such as Loss of Signal and is used to notify downstream
nodes (of the appropriate layer processing) that a failure has nodes (of the appropriate layer processing) that a failure has
occurred. AIS performs two functions 1) inform the intermediate occurred. AIS performs two functions 1) inform the intermediate
nodes (with the appropriate layer monitoring capability) that a nodes (with the appropriate layer monitoring capability) that a
failure has been detected 2) notify the connection end-point that failure has been detected 2) notify the connection end-point that
the service is no longer available. the service is no longer available.
For a distributed control plane supporting one (or more) failure For a distributed control plane supporting one (or more) failure
notification mechanism(s), regardless of the mechanismĂs actual notification mechanism(s), regardless of the mechanism's actual
implementation, the same capabilities are needed with more (or less) implementation, the same capabilities are needed with more (or less)
information provided about the LSPs/Spans under failure condition, information provided about the LSPs/spans under failure condition,
their detailed status, etc. their detailed status, etc.
The most important difference between these mechanisms is related to The most important difference between these mechanisms is related to
the fact that transport plane notifications (as defined today) would the fact that transport plane notifications (as defined today) would
initiate a protection scheme directly (such as those defined in directly initiate a protection type (such as those defined in
[CCAMP-TERM]) or a restoration scheme via the management plane. On [CCAMP-TERM]) via the transport plane or a restoration type/scheme
the other hand, using a failure notification mechanism through the via the management plane. The difference between recovery type and
control plane would provide the possibility to trigger either a scheme is detailed in Section 5.4.
On the other hand, using a failure notification mechanism through
the control plane would provide the possibility to trigger either a
protection or a restoration action via the control plane. This has protection or a restoration action via the control plane. This has
the advantage that a control plane recovery responsible entity does the advantage that a control plane recovery responsible entity does
not necessarily have to be co-located with a transport not necessarily have to be co-located with a transport
maintenance/recovery domain. A control plane recovery domain can be maintenance/recovery domain. A control plane recovery domain can be
defined at entities not supporting a transport plane recovery. defined at entities not supporting a transport plane recovery.
Moreover, as specified in [GMPLS-SIG], notification message Moreover, as specified in [RFC-3471], notification message exchanges
exchanges through a GMPLS control plane may not follow the same path through a GMPLS control plane may not follow the same path as the
as the LSP/spans for which these messages carry the status. In turn, LSP/spans for which these messages carry the status. In turn, this
this ensures a fast, reliable (through the use of either a dedicated ensures a fast, reliable (through acknowledgement and the use of
control plane network or disjoint control channels) and efficient either a dedicated control plane network or disjoint control
(through the aggregation of several LSP/span status within the same channels) and efficient (through the aggregation of several LSP/span
message) failure notification mechanism. status within the same message) failure notification mechanism.
The other important properties to be met by the failure notification The other important properties to be met by the failure notification
mechanism are mainly the following: mechanism are mainly the following:
- Notification messages must provide enough information such that - Notification messages must provide enough information such that
the most efficient subsequent recovery action will be taken (in the most efficient subsequent recovery action will be taken (in
most of the recovery schemes this action is even deterministic)
at the recovering entities. Remember here that these entities can D.Papadimitriou et al. - Internet Draft - Expires November 2003 7
be either intermediate or end-points through which normal traffic most of the recovery types and schemes this action is even
flows. Based on local policy, intermediate nodes may not use this deterministic) at the recovering entities. Remember here that
information for subsequent recovery actions (see for instance the these entities can be either intermediate or end-points through
APS protocol phases as described in [CCAMP-TERM]). In addition, which normal traffic flows. Based on local policy, intermediate
since fast notification is a mechanism running in collaboration nodes may not use this information for subsequent recovery actions
with the existing signalling (see for instance, [GMPLS-RSVP-TE]) (see for instance the APS protocol phases as described in [CCAMP-
allowing intermediate nodes to stay informed about the status of TERM]). In addition, since fast notification is a mechanism
the working LSP/spans under failure condition. running in collaboration with the existing signalling (see for
instance, [RFC-3473]), it allows intermediate nodes to stay
informed about the status of the working LSP/spans under failure
condition.
The trade-off here is to define what information the LSP/span end- The trade-off here is to define what information the LSP/span end-
points (more precisely, the deciding entity) needs in order for points (more precisely, the deciding entity) needs in order for
D.Papadimitriou et al. - Internet Draft ű June 2003 7
the recovering entity to take the best recovery action: if not the recovering entity to take the best recovery action: if not
enough information is provided, the decision can not be optimal enough information is provided, the decision can not be optimal
(note that in this eventuality, the important issue is to quantify (note that in this eventuality, the important issue is to quantify
the level of sub-optimality), if too much information is provided the level of sub-optimality), if too much information is provided
the control plane may be overloaded with unnecessary information the control plane may be overloaded with unnecessary information
and the aggregation/correlation of this notification information and the aggregation/correlation of this notification information
will be more complex and time consuming to achieve. Notice that will be more complex and time consuming to achieve. Note that a
more detailed quantification of the amount of information to be more detailed quantification of the amount of information to be
exchanged and processed is strongly dependent on the failure exchanged and processed is strongly dependent on the failure
notification protocol specification. notification protocol.
- If the failure localization and isolation is not performed by one - If the failure localization and isolation is not performed by one
of the LSP/Span end-points or some intermediate points, they of the LSP/span end-points or some intermediate points, they
should receive enough information from the notification message in should receive enough information from the notification message in
order to locate the failure otherwise they would need to (re-) order to locate the failure otherwise they would need to (re-)
initiate a failure localization and isolation action. initiate a failure localization and isolation action.
- Avoiding so-called notification storms implies that failure - Avoiding so-called notification storms implies that 1) the failure
detection output is correlated (i.e. alarm correlation) and detection output is correlated (i.e. alarm correlation) and
aggregated at the node detecting the failure(s), failure aggregated at the node detecting the failure(s) 2) the failure
notifications are directed to a restricted set of destinations (in notifications are directed to a restricted set of destinations (in
general the end-points) and notification suppression (i.e. alarm general the end-points) and that 3) failure notification
suppression) is provided in order to limit flooding in case of suppression (i.e. alarm suppression) is provided in order to limit
multiple and/or correlated failures appearing at several locations flooding in case of multiple and/or correlated failures appearing
in the network at several locations in the network.
- Alarm correlation and aggregation (at the failure detecting - Alarm correlation and aggregation (at the failure detecting
node) implies a consistent decision based on the conditions for node) implies a consistent decision based on the conditions for
which a trade-off between fast convergence (at detecting node) and which a trade-off between fast convergence (at detecting node) and
fast notification (implying that correlation and aggregation fast notification (implying that correlation and aggregation
occurs at receiving end-points) can be found. occurs at receiving end-points) can be found.
4.5 Correlating Failure Conditions 4.5 Correlating Failure Conditions
A single failure event (such as a span failure) can result into A single failure event (such as a span failure) can result into
reporting multiple failures (such as individual LSP failures) reporting multiple failures (such as individual LSP failures)
conditions. These can be grouped (i.e. correlated) to reduce the conditions. These can be grouped (i.e. correlated) to reduce the
number of failure conditions communicated on the reporting channel, number of failure conditions communicated on the reporting channel,
for both in-band and out-of-band failure reporting. for both in-band and out-of-band failure reporting.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 8
In such a scenario, it can be important to wait for a certain period In such a scenario, it can be important to wait for a certain period
of time, typically called failure correlation time, and gather all of time, typically called failure correlation time, and gather all
the failures to report them as a group of failures (or simply group the failures to report them as a group of failures (or simply group
failure). For instance, this approach can be provided using LMP-WDM failure). For instance, this approach can be provided using LMP-WDM
for pre-OTN networks (see [LMP-WDM]) or when using Signal Failure/ for pre-OTN networks (see [LMP-WDM]) or when using Signal Failure/
Degrade Group in the Sonet/SDH context. Degrade Group in the Sonet/SDH context.
Note that a default average time interval during which failure Note that a default average time interval during which failure
correlation operation can be performed is difficult to provide since correlation operation can be performed is difficult to provide since
it is strongly dependent on the underlying network topology. it is strongly dependent on the underlying network topology.
Therefore, it can be advisable to provide a per node configurable Therefore, it can be advisable to provide a per node configurable
failure correlation time. The detailed selection criteria for this failure correlation time. The detailed selection criteria for this
time interval are outside of the scope of this document. time interval are outside of the scope of this document.
D.Papadimitriou et al. - Internet Draft ű June 2003 8
When failure correlation is not provided, multiple failure When failure correlation is not provided, multiple failure
notification messages may be sent out in response to a single notification messages may be sent out in response to a single
failure (for instance, a fiber cut), each one containing a set of failure (for instance, a fiber cut), each one containing a set of
information on the failed working resources (for instance, the information on the failed working resources (for instance, the
individual lambda LSP flowing through this fiber). This allows for a individual lambda LSP flowing through this fiber). This allows for a
more prompt response but can potentially overload the control plane more prompt response but can potentially overload the control plane
due to a large amount of failure notifications. due to a large amount of failure notifications.
5. Recovery Mechanisms and Schemes 5. Recovery Mechanisms
5.1 Transport vs. Control Plane Responsibilities 5.1 Transport vs. Control Plane Responsibilities
For both protection and restoration, and when applicable, recovery For both protection and restoration, and when applicable, recovery
resources are provisioned using GMPLS signalling capabilities. Thus, resources are provisioned using GMPLS signalling capabilities. Thus,
these are control plane-driven actions (topological and resource- these are control plane-driven actions (topological and resource-
constrained) that are always performed in this context. constrained) that are always performed in this context.
The following table gives an overview of the responsibilities taken The following table gives an overview of the responsibilities taken
by the control plane in case of LSP/Span recovery: by the control plane in case of LSP/span recovery:
1. LSP/span Protection Schemes 1. LSP/span Protection Schemes
- Phase 1: Failure detection Transport plane - Phase 1: Failure detection Transport plane
- Phase 2: Failure isolation/localization Transport/Control plane - Phase 2: Failure localization/isolation Transport/Control plane
- Phase 3: Failure notification Transport/Control plane - Phase 3: Failure notification Transport/Control plane
- Phase 4: Protection switching Transport/Control plane - Phase 4: Protection switching Transport/Control plane
- Phase 5: Reversion (normalization) Transport/Control plane - Phase 5: Reversion (normalization) Transport/Control plane
Note: in the LSP/span protection context control plane actions can Note: in the LSP/span protection context control plane actions can
be performed either for operational purposes and/or synchronization be performed either for operational purposes and/or synchronization
purposes (vertical synchronization between transport and control purposes (vertical synchronization between transport and control
plane) and/or notification purposes (horizontal synchronization plane) and/or notification purposes (horizontal synchronization
between nodes at control plane level). between nodes at control plane level). This suggests the selection
of the responsible plane (in particular for protection switching)
during the provisioning phase of the protected/protection LSP.
2. LSP/span Restoration Schemes 2. LSP/span Restoration Schemes
D.Papadimitriou et al. - Internet Draft - Expires November 2003 9
- Phase 1: Failure detection Transport plane - Phase 1: Failure detection Transport plane
- Phase 2: Failure isolation/localization Transport/Control plane - Phase 2: Failure localization/isolation Transport/Control plane
- Phase 3: Failure notification Control plane - Phase 3: Failure notification Control plane
- Phase 4: Recovery switching Control plane - Phase 4: Recovery switching Control plane
- Phase 5: Reversion (normalization) Control plane - Phase 5: Reversion (normalization) Control plane
Therefore, this document is primarily focused on provisioning of Therefore, this document primarily focuses on provisioning of LSP
recovery resources, failure notification, LSP/span recovery and recovery resources, failure notification mechanisms, recovery
reversion operations. Moreover some additional considerations can be switching, and reversion operations. Moreover some additional
dedicated to the mechanisms associated to the failure localization/ considerations can be dedicated to the mechanisms associated to the
isolation phase. failure localization/isolation phase.
5.2 Technology in/dependent mechanisms 5.2 Technology in/dependent mechanisms
The present recovery mechanisms analysis applies in fact to any The present recovery mechanisms analysis applies in fact to any
circuit oriented data plane technology with discrete bandwidth circuit oriented data plane technology with discrete bandwidth
increments (like Sonet/SDH, G.709 OTN, etc.) being controlled by a
D.Papadimitriou et al. - Internet Draft ű June 2003 9 GMPLS-based distributed control plane.
increments (like Sonet/SDH, G.709 OTN, etc.) being controlled by an
IP-centric distributed control plane.
The following sub-sections are not intended to favor one technology The following sub-sections are not intended to favor one technology
versus another. They just lists pro and cons for each of them in versus another. They just lists pro and cons for each of them in
order to determine the mechanisms that GMPLS-based recovery must order to determine the mechanisms that GMPLS-based recovery must
deliver to overcome their cons and take benefits of their pros in deliver to overcome their cons and take benefits of their pros in
their respective applicability context. their respective applicability context.
5.2.1 OTN Recovery 5.2.1 OTN Recovery
OTN recovery specifics are left for further considerations. OTN recovery specifics are left for further considerations.
5.2.2 Pre-OTN Recovery 5.2.2 Pre-OTN Recovery
Pre-OTN Recovery specifics (also referred to as ˘lambda switching÷) Pre-OTN recovery specifics (also referred to as "lambda switching")
presents mainly the following advantages: presents mainly the following advantages:
- benefits from a simpler architecture making it more suitable for - benefits from a simpler architecture making it more suitable for
meshed-based recovery schemes (on a per channel basis). mesh-based recovery types and schemes (on a per channel basis).
- when providing suppression of intermediate node transponders (vs. - when providing suppression of intermediate node transponders (vs.
use of non-standard masking of upstream failures) e.g. use of use of non-standard masking of upstream failures) e.g. use of
squelching, implies that failures (such as LoL) will propagate to squelching, implies that failures (such as LoL) will propagate to
edge nodes giving the possibility to initiate upper layer driven edge nodes giving the possibility to initiate upper layer driven
recovery actions. recovery actions.
The main disadvantage comes from the lack of interworking due to the The main disadvantage comes from the lack of interworking due to the
large amount of failure management (in particular failure large amount of failure management (in particular failure
notification protocols) and recovery mechanisms currently available. notification protocols) and recovery mechanisms currently available.
Note also, that for all-optical networks, combination of recovery Note also, that for all-optical networks, combination of recovery
with optical physical impairments is left for a future release of with optical physical impairments is left for a future release of
this document since corresponding detection technologies are under this document since corresponding detection technologies are under
specification. specification.
5.2.3 Sonet/SDH Recovery 5.2.3 Sonet/SDH Recovery
D.Papadimitriou et al. - Internet Draft - Expires November 2003 10
Some of the advantages of Sonet/SDH and more generically any TDM Some of the advantages of Sonet/SDH and more generically any TDM
transport plane are: transport plane recovery are that they provide:
- Protection schemes are standardized (see [G.841]) and can operate - Protection types operating at the data plane level are
across protected domains and interwork (see [G.842]). standardized (see [G.841]) and can operate across protected
domains and interwork (see [G.842]).
- Provides failure detection, notification and path/section - Failure detection, notification and path/section Automatic
Automatic Protection Switching (APS) mechanisms. Protection Switching (APS) mechanisms.
- Provides greater control over the granularity of the TDM - Greater control over the granularity of the TDM LSPs/links that
LSPs/Links that can be recovered with respect to coarser optical can be recovered with respect to coarser optical channel (or whole
channel (or whole fiber content) recovery switching fiber content) recovery switching
Some of the current limitations of the Sonet/SDH layer recovery are: Some of the limitations of the Sonet/SDH layer recovery are:
D.Papadimitriou et al. - Internet Draft ű June 2003 10
- Limited topological scope: Inherently the use of ring topologies - Limited topological scope: Inherently the use of ring topologies
(Dedicated SNCP or Shared Protection Rings) has a reduced (Dedicated SNCP or Shared Protection Rings) has a reduced
flexibility with respect to the somewhat more complex but flexibility with respect to the somewhat more complex and
potentially more resource efficient mesh-based recovery schemes. more resource efficient mesh-based recovery types and schemes.
- Inefficient use of spare capacity: Sonet/SDH protection is largely - Inefficient use of spare capacity: Sonet/SDH protection is largely
applied for ring topologies, where spare capacity often remains applied for ring topologies, where spare capacity often remains
idle, making the efficiency of bandwidth usage an issue. idle, making the efficiency of bandwidth usage an issue.
- Support of meshed recovery requires intensive network management - Support of meshed recovery requires intensive network management
development, and the functionality is limited by both the network development and the functionality is limited by both the network
elements and the element management systems capabilities. elements and the element management systems capabilities.
5.3 Specific Aspects of Control Plane-based Recovery Mechanisms 5.3 Specific Aspects of Control Plane-based Recovery Mechanisms
5.3.1 In-band vs Out-of-band Signalling 5.3.1 In-band vs Out-of-band Signalling
The nodes communicate through the use of (IP terminating) control The nodes communicate through the use of IP terminating control
channels defining the control plane (transport) topology. In this channels defining the control plane (transport) topology. In this
context, two classes of transport mechanisms can be considered here context, two classes of transport mechanisms can be considered here
i.e. in-fiber or out-of-fiber (through a dedicated physically i.e. in-fiber or out-of-fiber (through a dedicated physically
diverse control network referred to as the Data Communication diverse control network referred to as the Data Communication
Network or DCN). The potential impact of the usage of an in-fiber Network or DCN). The potential impact of the usage of an in-fiber
(signalling) transport mechanism is briefly considered here. (signalling) transport mechanism is briefly considered here.
In-fiber transport mechanism can be further subdivided into in-band In-fiber transport mechanism can be further subdivided into in-band
and out-of-band. As such, the distinction between in-fiber in-band and out-of-band. As such, the distinction between in-fiber in-band
and in-fiber out-of-band signalling reduces to the consideration of and in-fiber out-of-band signalling reduces to the consideration of
a logically versus physically embedded control plane topology with a logically versus physically embedded control plane topology with
respect to the transport plane topology. In the scope of this respect to the transport plane topology. In the scope of this
document, since we assume that (IP terminating) channels between document, since we assume that IP terminating control channels
nodes must be continuously available in order to enable the exchange between nodes must be continuously available to enable the exchange
of recovery-related information and messages, one considers that in of recovery-related information and messages, one considers that in
either case (i.e. in-band or out-of-band) at least one logical either case (i.e. in-band or out-of-band) at least one logical
channel or one physical channel between nodes is available. channel or one physical channel between nodes is always available.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 11
Therefore, the key issue when using in-fiber signalling is whether Therefore, the key issue when using in-fiber signalling is whether
we can assume independence between the fault-tolerance capabilities we can assume independence between the fault-tolerance capabilities
of control plane and the failures affecting the transport plane of control plane and the failures affecting the transport plane
(including the nodes). Note also that existing specifications like (including the nodes). Note also that existing specifications like
the OTN provide a limited form of independence for in-fiber the OTN provide a limited form of independence for in-fiber
signaling by dedicating a separate optical supervisory channel (OSC, signaling by dedicating a separate optical supervisory channel (OSC,
see [ITU-T G.709] and [ITU-T G.874]) to transport the overhead and see [G.709] and [G.874]) to transport the overhead and other control
other control traffic. For OTNs, failure of the OSC does not result traffic. For OTNs, failure of the OSC does not result in failing the
in failing the optical channels. Similarly, loss of the control optical channels. Similarly, loss of the control channel must not
channel must not result in failing the data (transport plane). result in failing the data channels (transport plane).
5.3.2 Uni- versus Bi-directional Failures 5.3.2 Uni- versus Bi-directional Failures
The failure detection, correlation and notification mechanisms The failure detection, correlation and notification mechanisms
(described in Section 4) can be triggered when either a (described in Section 4) can be triggered when either a
unidirectional or a bi-directional LSP/Span failure occurs (or a unidirectional or a bi-directional LSP/Span failure occurs (or a
D.Papadimitriou et al. - Internet Draft ű June 2003 11
combination of both). As illustrated in Figure 1 and 2, two combination of both). As illustrated in Figure 1 and 2, two
alternatives can be considered here: alternatives can be considered here:
1. Uni-directional failure detection: the failure is detected on the 1. Uni-directional failure detection: the failure is detected on the
receiver side i.e. it is only is detected by the downstream node receiver side i.e. it is only is detected by the downstream node
to the failure (or by the upstream node depending on the failure to the failure (or by the upstream node depending on the failure
propagation direction, respectively) propagation direction, respectively).
2. Bi-directional failure detection: the failure is detected on the 2. Bi-directional failure detection: the failure is detected on the
receiver side of both downstream node AND upstream node to the receiver side of both downstream node AND upstream node to the
failure. failure.
Notice that after the failure detection time, if only control plane Notice that after the failure detection time, if only control plane
based failure management is provided, the peering node is unaware of based failure management is provided, the peering node is unaware of
the failure detection status of its neighbor. the failure detection status of its neighbor.
------- ------- ------- ------- ------- ------- ------- -------
skipping to change at line 629 skipping to change at line 645
Up Notification Down Notification Up Notification Down Notification
------- ------- ------- ------- ------- ------- ------- -------
| | | |Tx Rx| | | | | | | |Tx Rx| | | |
| NodeA |----...----| NodeB |xxxxxxxxx| NodeC |----...----| NodeD | | NodeA |----...----| NodeB |xxxxxxxxx| NodeC |----...----| NodeD |
| |----...----| |xxxxxxxxx| |----...----| | | |----...----| |xxxxxxxxx| |----...----| |
------- ------- ------- ------- ------- ------- ------- -------
t0 F <<<<<<< >>>>>>> F t0 F <<<<<<< >>>>>>> F
D.Papadimitriou et al. - Internet Draft - Expires November 2003 12
t1 x <-------------> x t1 x <-------------> x
Notification Notification
t2 <--------...--------x x--------...--------> t2 <--------...--------x x--------...-------->
Up Notification Down Notification Up Notification Down Notification
Fig. 1 & 2. Uni- and Bi-directional Failure Detection/Notification
After failure detection, the following failure management operations After failure detection, the following failure management operations
can be subsequently considered: can be subsequently considered:
- Each detecting entity sends a notification message to the - Each detecting entity sends a notification message to the
corresponding transmitting entity. For instance, in Fig. 1 (Fig. corresponding transmitting entity. For instance, in Fig. 1 (Fig.
2), node C sends a notification message to node B (while node B 2), node C sends a notification message to node B (while node B
D.Papadimitriou et al. - Internet Draft ű June 2003 12
sends a notification message to node A). To ensure reliable sends a notification message to node A). To ensure reliable
failure notification, a dedicated acknowledgment message can be failure notification, a dedicated acknowledgment message can be
returned back to the sender node. returned back to the sender node.
- Next, within a certain (and pre-determined) time window, nodes - Next, within a certain (and pre-determined) time window, nodes
impacted by the failure occurrences perform their correlation. In impacted by the failure occurrences perform their correlation. In
case of unidirectional failure, node B only receives the case of unidirectional failure, node B only receives the
notification message from node C and thus the time for this notification message from node C and thus the time for this
operation is negligible. However, in case of bi-directional operation is negligible. However, in case of bi-directional
failure, node B (and node C) must correlate the received failure, node B (and node C) must correlate the received
skipping to change at line 676 skipping to change at line 689
action. Note that the connection terminating node (i.e. node D or action. Note that the connection terminating node (i.e. node D or
node A) may be optionally notified. node A) may be optionally notified.
In case of bi-directional failure, node B may send an upstream In case of bi-directional failure, node B may send an upstream
notification message to the ingress node A or node C a downstream notification message to the ingress node A or node C a downstream
notification to the egress node D. However, due to the dependence notification to the egress node D. However, due to the dependence
on the connection directionality, only ingress node A or egress on the connection directionality, only ingress node A or egress
node D would initiate an edge to edge recovery action. Note that node D would initiate an edge to edge recovery action. Note that
the connection terminating node (i.e. node D or node A) should be the connection terminating node (i.e. node D or node A) should be
also notified of this event using upstream and downstream fast also notified of this event using upstream and downstream fast
notification (see [GMPLS-SIG]). For instance, if a connection notification (see [RFC-3471]). For instance, if a connection
directed from D to A is under failure condition, only the directed from D to A is under failure condition, only the
notification sent by from node C to D would initiate a recovery notification sent from node C to D would initiate a recovery
action. Here as well, per [CCAMP-TERM], the deciding (and action. Here as well, per [CCAMP-TERM], the deciding (and
recovering) node D is referred to as the "master" while the node A recovering) node D is referred to as the "master" while the node A
is referred to as the "slave" (i.e. recovering only entity). is referred to as the "slave" (i.e. recovering only entity).
Note: The determination of the master and the slave may be based Note: The determination of the master and the slave may be based
D.Papadimitriou et al. - Internet Draft - Expires November 2003 13
either on configured information or dedicated protocol capability. either on configured information or dedicated protocol capability.
In the above scenarios, the path followed by the notification In the above scenarios, the path followed by the notification
messages does not have to be the same as the one followed by the messages does not have to be the same as the one followed by the
failed LSP (see [GMPLS-SIG], for more details on the notification failed LSP (see [RFC-3471], for more details on the notification
message exchange). The important point, concerning this mechanism, message exchange). The important point, concerning this mechanism,
is that either the detecting/reporting entity (i.e. the nodes B and is that either the detecting/reporting entity (i.e. the nodes B and
C) are also the deciding/recovery entity or the detecting/reporting C) are also the deciding/recovery entity or the detecting/reporting
entities are simply intermediate nodes in the subsequent recovery entities are simply intermediate nodes in the subsequent recovery
process. One refers to local recovery in the former case and to process. One refers to local recovery in the former case and to
edge-to-edge recovery in the latter one. edge-to-edge recovery in the latter one.
5.3.3 Partial versus Full Span Recovery 5.3.3 Partial versus Full Span Recovery
D.Papadimitriou et al. - Internet Draft ű June 2003 13
When given span carries more than one LSPs or LSP segments, an When given span carries more than one LSPs or LSP segments, an
additional aspect must be considered during span failure carrying additional aspect must be considered during span failure carrying
several LSPs. These LSPs can be either individually recovered or several LSPs. These LSPs can be either individually recovered or
recovered as a group (aka bulk LSP recovery) or independent sub- recovered as a group (aka bulk LSP recovery) or independent sub-
groups. The selection of this mechanism would be triggered groups. The selection of this mechanism would be triggered
independently of the failure notification granularity when independently of the failure notification granularity when
correlation time windows are used and simultaneous recovery of correlation time windows are used and simultaneous recovery of
several LSPs can be performed using single request. Moreover, several LSPs can be performed using single request. Moreover,
criteria by which such sub-groups can be formed are outside of the criteria by which such sub-groups can be formed are outside of the
scope of this document. scope of this document.
skipping to change at line 738 skipping to change at line 752
The recovery definitions given in [CCAMP-TERM] are quite generic and The recovery definitions given in [CCAMP-TERM] are quite generic and
apply for link (or local span) and LSP recovery. The major apply for link (or local span) and LSP recovery. The major
difference between LSP, LSP Segment and span recovery is related to difference between LSP, LSP Segment and span recovery is related to
the number of intermediate nodes that the signalling messages have the number of intermediate nodes that the signalling messages have
to travel. Since nodes are not necessarily adjacent in case of LSP to travel. Since nodes are not necessarily adjacent in case of LSP
(or LSP Segment) recovery, signalling message exchanges from the (or LSP Segment) recovery, signalling message exchanges from the
reporting to the deciding/recovery entity will have to cross several reporting to the deciding/recovery entity will have to cross several
intermediate nodes. In particular, this applies for the notification intermediate nodes. In particular, this applies for the notification
messages due to the number of hops separating the failure occurrence messages due to the number of hops separating the failure occurrence
D.Papadimitriou et al. - Internet Draft - Expires November 2003 14
location from their destination. This results in an additional location from their destination. This results in an additional
propagation and forwarding delay. Note that the former delay may in propagation and forwarding delay. Note that the former delay may in
certain circumstances be non-negligible e.g. in case of copper out- certain circumstances be non-negligible e.g. in case of copper out-
of-band network one has to consider approximately 1 ms per 200km. of-band network one has to consider approximately 1 ms per 200km.
Moreover, the recovery mechanisms applicable to end-to-end LSP and Moreover, the recovery mechanisms applicable to end-to-end LSP and
to the segments (i.e. edge-to-edge) that may compose an end-to-end to the segments (i.e. edge-to-edge recovery) that may compose an
LSP can be exactly the same. However, one expects in the latter end-to-end LSP can be exactly the same. However, one expects in the
case, that the destination of the failure notification message will latter case, that the destination of the failure notification
be the ingress of each of these segments. Therefore, taking into message will be the ingress of each of these segments. Therefore,
account the mechanism described in Section 5.3.2, failure taking into account the mechanism described in Section 5.3.2,
notification can be first exchanged between the LSP segments failure notification can be first exchanged between the LSP segments
D.Papadimitriou et al. - Internet Draft ű June 2003 14
terminating points and after expiration of the hold-off time terminating points and after expiration of the hold-off time
directed toward end-to-end LSP terminating points. directed toward end-to-end LSP terminating points.
Note: Several studies provide quantitative analysis of the relative
performance of LSP/span recovery techniques. [WANG] for instance,
provides an analysis grid for these techniques showing that dynamic
LSP restoration (see Section 5.5.2) performs well under medium
network loads but suffers performance degradations at higher loads
due to greater contention for recovery resources. LSP restoration
upon span failure, as defined in [WANG], degrades at higher loads
because paths around failed links tend to increase the hop count of
the affected LSPs and thus consume additional network resources.
Also, LSP restoration's performance can be enhanced by a failed
working LSP's source node launching a new recovery attempt if an
initial attempt fails. A single retry attempt is sufficient to
produce large increases in restoration success rate and
availability, especially at high loads, while not adding
significantly to the long-term average recovery time. Allowing
additional attempts produces only small additional gains in
performance. This suggests using additional (intermediate) crankback
signalling when using dynamic LSP restoration (described in Section
5.5.2 - case 2). Details on crankback signalling are outside of
scope of the present document.
5.4 Difference between Recovery Type and Scheme 5.4 Difference between Recovery Type and Scheme
Section 4.6 of [CCAMP-TERM] defines the basic recovery types. The Section 4.6 of [CCAMP-TERM] defines the basic LSP/span recovery
purpose of this section is to describe the schemes that can be built types. The purpose of this section is to describe the recovery
using these recovery types. In brief, a recovery scheme is defined schemes that can be built using these recovery types. In brief, a
as the combination between different ingress-egress node pairs of a recovery scheme is defined as the combination of several ingress-
set of identical recovery types. Several examples are provided in egress node pairs supporting a given recovery type (from the set of
order to illustrate the difference between a recovery type such as the recovery types they allow). Several examples are provided here
1:1 and a recovery scheme such as (1:1)^n. to illustrate the difference between recovery types such as 1:1 or
M:N and recovery schemes such as (1:1)^n or (M:N)^n referred to as
shared-mesh recovery.
1. (1:1)^n with recovery resource sharing 1. (1:1)^n with recovery resource sharing
The exponent, n, indicates the number of times a 1:1 recovery type The exponent, n, indicates the number of times a 1:1 recovery type
is applied between at most n different ingress-egress node pairs. is applied between at most n different ingress-egress node pairs.
Here, at most n pairs of disjoint working and recovery LSPs/spans Here, at most n pairs of disjoint working and recovery LSPs/spans
D.Papadimitriou et al. - Internet Draft - Expires November 2003 15
share at most n times a common resource. Since the working LSPs/ share at most n times a common resource. Since the working LSPs/
spans are mutually disjoint, simultaneous requests for use of the spans are mutually disjoint, simultaneous requests for use of the
shared (common) resource will only occur in case of simultaneous shared (common) resource will only occur in case of simultaneous
failures, which is less likely to happen. failures, which is less likely to happen.
For instance, in the (1:1)^2 common case if the 2 recovery LSPs in For instance, in the (1:1)^2 common case if the 2 recovery LSPs in
the group overlap the same common resource, then it can handle only the group overlap the same common resource, then it can handle only
single failures; any multiple working LSP failures will cause at single failures; any multiple working LSP failures will cause at
least one working LSP to be denied automatic recovery. Consider for least one working LSP to be denied automatic recovery. Consider for
instance, the following example, with working LSPs A-B and E-F and instance, the following topology, with working LSPs A-B-C and F-G-H
recovery LSPs A-C-D-B and E-C-D-F sharing a common C-D resource. and recovery LSPs A-D-E-C and F-D-E-H sharing a common D-E link
resource.
A ----------------- B A---------B---------C
\ / \ /
\ / \ /
C ----------- D D-------------E
/ \ / \
/ \ / \
E ----------------- F F---------G---------H
2. (M:N)^n with recovery resource sharing 2. (M:N)^n with recovery resource sharing
The exponent, n, indicates the number of times a M:N recovery type The (M:N)^n scheme is documented here for the sake of completeness
is applied between at most n different ingress-egress node pairs. only (i.e. it is not expected that GMPLS capabilities would support
So the interpretation follows from the previous case, expect that this scheme). The exponent, n, indicates the number of times a M:N
here disjointness applies to the N working LSPs/spans and to the M recovery type is applied between at most n different ingress-egress
recovery LSPs/spans while sharing at most n times M common node pairs. So the interpretation follows from the previous case,
resources. expect that here disjointness applies to the N working LSPs/spans
and to the M recovery LSPs/spans while sharing at most n times M
common resources.
In both schemes, one may see the following at the LSP level: we have In both schemes, one may see the following at the LSP level: we have
a ˘group÷ of sum{n=1}^N N{n} working LSPs and a pool of shared a "group" of sum{n=1}^N N{n} working LSPs and a pool of shared
backup resources, not all of which are available to any given recovery resources, not all of which are available to any given
working path. In such conditions, defining a metric that describes working path. In such conditions, defining a metric that describes
the amount of overlap among the recovery LSPs would give some the amount of overlap among the recovery LSPs would give some
indication of the group's ability to handle multiple simultaneous
D.Papadimitriou et al. - Internet Draft ű June 2003 15
indication of the groupĂs ability to handle multiple simultaneous
failures. failures.
For instance, in the simple (1:1)^n case situation if n recovery For instance, in the simple (1:1)^n case situation if n recovery
LSPs in a (1:1)^n group overlap, then it can handle only single LSPs in a (1:1)^n group overlap, then it can handle only single
failures; any multiple working LSP failures will cause at least one failures; any multiple working LSP failures will cause at least one
working LSP to be denied automatic recovery. But if one consider for working LSP to be denied automatic recovery. But if one consider for
instance, a (2:2)^2 group in which there are two pairs of instance, a (2:2)^2 group in which there are two pairs of
overlapping recovery LSPs, then two LSPs (belonging to the same overlapping recovery LSPs, then two LSPs (belonging to the same
pair) can be simultaneously recovered. The latter case can be pair) can be simultaneously recovered. The latter case can be
illustrated as follows: 2 working LSPs A-B and E-F and 2 recovery illustrated as follows: 2 working LSPs A-B-C and F-G-H and 2
LSPs A-C-D-B and E-C-D-F sharing the two common C-D resources. recovery LSPs A-D-E-C and F-D-E-H sharing the two common D-E link
resources.
A ================ B D.Papadimitriou et al. - Internet Draft - Expires November 2003 16
A========B========C
\\ // \\ //
\\ // \\ //
C =========== D D =========== E
// \\ // \\
// \\ // \\
E ================ F F========G========H
Moreover, in all these schemes, (working) path disjointness can be Moreover, in all these schemes, (working) path disjointness can be
reinforced by exchanging working LSP related information during the reinforced by exchanging working LSP related information during the
recovery LSP signalling. Specific issues related to the combination recovery LSP signalling. Specific issues related to the combination
of shared (discrete) bandwidth and disjointness for recovery schemes of shared (discrete) bandwidth and disjointness for recovery schemes
are described in Section 8.4.2. are described in Section 8.4.2.
5.5 LSP Restoration Schemes 5.5 LSP Recovery Mechanisms
5.5.1 Classification 5.5.1 Classification
LSPs/spans recovery time and ratio depend on the proper recovery LSP LSPs/spans recovery time and ratio depend on the proper recovery LSP
(soft) provisioning and the level of recovery resources overbooking provisioning (meaning pre-provisioning when performed before failure
(i.e. over-provisioning). A proper balance of these two mechanisms occurrence) and the level of recovery resources overbooking (i.e.
will result in a desired LSP/span recovery time and ratio when over-provisioning). A proper balance of these two mechanisms will
single or multiple failure(s) occur(s). result in a desired LSP/span recovery time and ratio when single or
multiple failure(s) occur(s).
Recovery LSP Provisioning phases: Recovery LSP provisioning phases:
(1) Route Computation --> On-demand (1) Route Computation --> On-demand
| |
| |
--> Pre-Computed --> Pre-Computed
| |
| |
(2) Signalling --> On-demand (2) Signalling --> On-demand
| |
| |
--> Pre-Signaled --> Pre-Signaled
| |
| |
(3) Resource Selection --> On-demand (3) Resource Selection --> On-demand
D.Papadimitriou et al. - Internet Draft ű June 2003 16
| |
| |
--> Pre-Selected --> Pre-Selected
Overbooking levels:
Overbooking Levels:
+----- Dedicated (for instance: 1+1, 1:1, etc.) +----- Dedicated (for instance: 1+1, 1:1, etc.)
| |
| |
+----- Shared (for instance: 1:N, M:N, etc.) +----- Shared (for instance: 1:N, M:N, etc.)
| |
Level of | Level of |
Overbooking -----+----- Unprotected (for instance: 0:1, 0:N) Overbooking -----+----- Unprotected (for instance: 0:1, 0:N)
Fig 3. LSP Provisioning and Overbooking Classification D.Papadimitriou et al. - Internet Draft - Expires November 2003 17
In this figure, we present a classification of different options In this figure, we present a classification of different options
under LSP provisioning and overbooking. Although we acknowledge under LSP (pre-)provisioning and overbooking. Although these
these operations are run mostly during planning (using network operations are mostly performed during network planning and (pre-)
planning) and provisioning time (using signaling and routing) provisioning phases using GMPLS signaling capabilities, we keep them
activities, we keep them in analyzing the recovery schemes. in analyzing the recovery types.
Proper LSP/span provisioning will help in alleviating many of the Proper LSP/span (pre-)provisioning will help in alleviating many of
failures. As an example, one may compute primary and secondary the failures. As an example, one may compute and establish the
paths, either end-to-end or segment-per-segment, to recover an LSP working and the recovery paths either end-to-end or segment-per-
from multiple failure events affecting link(s), node(s), SRLG(s) segment, to protect an LSP from multiple failure events affecting
and/or SRG(s). Such primary and secondary LSP/span provisioning can link(s), node(s) and/or SRLG(s). Such working and recovery LSP/span
be categorized, as shown in the above figure, based on: provisioning can be categorized, as shown in the above figure, as
follows:
(1) the recovery path (i.e. route) can be either pre-computed or (1) the recovery path (i.e. route) can be either pre-computed or
computed on demand. computed on demand.
(2) when the recovery path is pre-computed: pre-signaled (implying (2) when the recovery path is pre-computed: pre-signaled (implying
recovery resource reservation) or signaled on demand. recovery resource reservation) or signaled on demand.
(3) and when the recovery resources are reserved, they can be either (3) and when the recovery resources are pre-signaled, they can be
pre-selected or selection on-demand. either pre-selected or selected on-demand.
Note that these different options give rise to different LSP/span Note that these different options give rise to different LSP/span
recovery times. The following subsections will consider all these recovery times. The following subsections will consider all the
cases in analyzing the schemes. above-mentioned (pre-)provisioning scenarios when analyzing the
different recovery mechanisms.
There are many mechanisms available allowing the overbooking of the There are many mechanisms available allowing the overbooking of the
recovery resources. This overbooking can be done per LSP (such as recovery resources. This overbooking can be done per LSP (such as
the example mentioned above), per link (such as span protection) or the example mentioned above), per link (such as span protection) or
per domain (such as ring topologies). In all these cases the level per domain (such as ring topologies). In all these cases the level
of overbooking, as shown in the above figure, can be classified as of overbooking, as shown in the above figure, can be classified as
dedicated (such as 1+1 and 1:1), shared (such as 1:N and M:N) or dedicated (such as 1+1 and 1:1), shared (such as 1:N and M:N) or
unprotected (i.e. restorable if enough recovery resources are unprotected (i.e. restorable if enough recovery resources are
available). available).
Under a shared restoration scheme one may support preemptable When using shared restoration, one may support preemptable (preempt
(preempt low priority connections in case of resource contention) low priority connections in case of resource contention) extra-
traffic. In this document, we consider all the above-mentioned
overbooking mechanisms in analyzing the corresponding recovery
scheme.
D.Papadimitriou et al. - Internet Draft ű June 2003 17 5.5.2 LSP Restoration Mechanisms
extra-traffic. In this document we keep in mind all the above-
mentioned overbooking mechanisms in analyzing the recovery schemes.
5.5.2 Dynamic LSP Restoration First, we define the following times to provide a quantitative
estimation about the time performance of the different LSP
restoration mechanisms (also referred to as LSP re-routing):
We first define the following times in order to provide a - Path Computation Time: Tc
quantitative estimation about the time performance of the different - Path Selection Time: Ts
dynamic and pre-signaled LSP restoration schemes (note: restoration - End-to-end LSP Resource Reservation: Tr (a delta for resource
is also referred to as re-routing):
- Path Computation Time - Tpc
- Path Selection Time - Tps
- End-to-end LSP Resource Reservation ű Trr (a delta for resource
selection is also considered, the corresponding total time is then selection is also considered, the corresponding total time is then
referred to as Trrs) referred to as Trs)
- End-to-end LSP Resource Activation Time ű Tra (a delta for - End-to-end LSP Resource Activation Time: Ta (a delta for
D.Papadimitriou et al. - Internet Draft - Expires November 2003 18
resource selection is also considered, the corresponding total resource selection is also considered, the corresponding total
time is then referred to as Tras) time is then referred to as Tas)
Path Selection Time (Tps) is considered when a pool of recovery The Path Selection Time (Ts) is considered when a pool of recovery
LSPĂs paths between a given source/destination is pre-computed and LSPs paths between a given source/destination is pre-computed and
after failure occurrence one of these paths is selected for the after failure occurrence one of these paths is selected for the
recovery of the LSP under failure condition. recovery of the LSP under failure condition.
Note: failure management operations such as failure detection, Note: failure management operations such as failure detection,
correlation and notification are considered as equivalently time correlation and notification are considered as equivalently time
consuming for all the mechanisms described here below: consuming for all the mechanisms described here below:
1. With Route Pre-computation (or LSP re-provisioning) 1. With Route Pre-computation (or LSP re-provisioning)
An end-to-end restoration LSP is established after the failure(s) An end-to-end restoration LSP is established after the failure(s)
occur(s) based on a pre-computed path (i.e. route). As such, one can occur(s) based on a pre-computed path (i.e. route). As such, one can
define this as an ˘LSP re-provisioning÷ mechanism. Here, one or more define this as an "LSP re-provisioning" mechanism. Here, one or more
(disjoint) routes for the restoration path are computed (and (disjoint) routes for the restoration path are computed (and
optionally pre-selected) before a failure occurs. optionally pre-selected) before a failure occurs.
No reservation or selection of resources is performed along the No reservation or selection of resources is performed along the
restoration path before failure. As a result, there is no guarantee restoration path before failure. As a result, there is no guarantee
that a restoration connection is available when a failure occurs. that a restoration connection is available when a failure occurs.
The expected total restoration time T is thus equal to Tps + Trrs or The expected total restoration time T is thus equal to Ts + Trs or
when a dedicated computation is performed for each working LSP to when a dedicated computation is performed for each working LSP to
Trrs. Trs.
2. Without Route Pre-computation (or LSP re-routing) 2. Without Route Pre-computation (or Full LSP re-routing)
An end-to-end restoration LSP is established after the failure(s) An end-to-end restoration LSP is dynamically established after the
occur(s). Here, one or more (disjoint) explicit routes for the failure(s) occur(s). Here, one or more (disjoint) explicit routes
restoration path are dynamically computed and one is selected after for the restoration path are dynamically computed and one is
failure. As such, one can define this as an ˘LSP re-routing÷ selected after failure. As such, one can define this as a complete
mechanism. "LSP re-routing" mechanism.
D.Papadimitriou et al. - Internet Draft ű June 2003 18
No reservation or selection of resources is performed along the No reservation or selection of resources is performed along the
restoration path before failure. As a result, there is no guarantee restoration path before failure. As a result, there is no guarantee
that a restoration connection is available when a failure occurs. that a restoration connection is available when a failure occurs.
The expected total restoration time T is thus equal to Tpc (+ Tps) + The expected total restoration time T is thus equal to Tc (+ Ts) +
Trrs. Therefore, time performance between these two approaches Trs. Therefore, time performance between these two approaches
differs by the time required for route computation Tpc (and its differs by the time required for route computation Tc (and its
potential selection time, Tps). potential selection time, Ts).
5.5.3 Pre-planned LSP Restoration 5.5.3 Pre-planned LSP Restoration
Pre-planned LSP restoration (also referred to as pre-planned LSP re- Pre-planned LSP restoration (also referred to as pre-planned LSP re-
routing) implies that the restoration LSP is pre-signaled. This in routing) implies that the restoration LSP is pre-signaled. This in
turn implies the reservation of recovery resources along the turn implies the reservation of recovery resources along the
restoration path. Two cases can be defined based on whether the restoration path. Two cases can be defined based on whether the
recovery resources are pre-selected or not. recovery resources are pre-selected or not.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 19
1. With resource reservation and without resource pre-selection 1. With resource reservation and without resource pre-selection
An end-to-end restoration path is pre-selected from a set of one or An end-to-end restoration path is pre-selected from a set of one or
more pre-computed (disjoint) explicit route before failure. The more pre-computed (disjoint) explicit route before failure. The
restoration LSP is signaled along this pre-selected path to reserve restoration LSP is signaled along this pre-selected path to reserve
resources (i.e. signaled) at each node but resources are not resources (i.e. signaled) at each node but resources are not
selected. selected.
In this case, the resources reserved for each restoration LSP may be In this case, the resources reserved for each restoration LSP may be
dedicated or shared between different working LSP that are not dedicated or shared between different working LSP that are not
expected to fail simultaneously. Local node policies can be applied expected to fail simultaneously. Local node policies can be applied
to define the degree to which these resources are shared across to define the degree to which these resources are shared across
independent failures. independent failures.
Upon failure detection, signaling is initiated along the restoration Upon failure detection, signaling is initiated along the restoration
path to select the resources, and to perform the appropriate path to select the resources, and to perform the appropriate
operation at each node entity involved in the restoration connection operation at each node entity involved in the restoration connection
(e.g. cross-connections). (e.g. cross-connections).
The expected total restoration time T is thus equal to Tras (post- The expected total restoration time T is thus equal to Tas (post-
failure activation) while operations performed before failure failure activation) while operations performed before failure
occurrence takes Tpc + Tps + Trr. occurrence takes Tc + Ts + Tr.
2. With both resource reservation and resource pre-selection 2. With both resource reservation and resource pre-selection
An end-to-end restoration path is pre-selected from a set of one or An end-to-end restoration path is pre-selected from a set of one or
more pre-computed (disjoint) explicit route before failure. The more pre-computed (disjoint) explicit route before failure. The
restoration LSP is signaled along this pre-selected path to reserve restoration LSP is signaled along this pre-selected path to reserve
AND select resources at each node but not cross-connected. Such that AND select resources at each node but not cross-connected. Such that
the selection of the recovery resources is fixed at the control the selection of the recovery resources is fixed at the control
plane level. However, no cross-connections are performed along the plane level. However, no cross-connections are performed along the
restoration path. restoration path.
In this case, the resources reserved for each restoration LSP may In this case, the resources reserved for each restoration LSP may
only be shared between different working LSPs that are not expected only be shared between different working LSPs that are not expected
to fail simultaneously. Since one considers restoration schemes to fail simultaneously. Since a restoration scheme is considered
D.Papadimitriou et al. - Internet Draft ű June 2003 19
here, the sharing degree should not be limited to working (and here, the sharing degree should not be limited to working (and
recovery) LSPs starting and ending at the same ingress and egress recovery) LSPs starting and ending at the same ingress and egress
nodes. Therefore, one expects to receive some feedback information nodes. Therefore, one expects to receive some feedback information
on the recovery resource sharing degree at each node participating on the recovery resource sharing degree at each node participating
to the recovery scheme. to the recovery scheme.
Upon failure detection, signaling is initiated along the restoration Upon failure detection, signaling is initiated along the restoration
path to activate the reserved and selected resources and to perform path to activate the reserved and selected resources and to perform
the appropriate operation at each node involved in the restoration the appropriate operation at each node involved in the restoration
connection (e.g. cross-connections). connection (e.g. cross-connections).
The expected total restoration time T is thus equal to Tra (post- The expected total restoration time T is thus equal to Ta (post-
failure activation) while operations performed before failure failure activation) while operations performed before failure
occurrence takes Tpc + Tps + Trrs. Therefore, time performance occurrence takes Tc + Ts + Trs. Therefore, time performance between
between these two approaches differs only by the time required for these two approaches differs only by the time required for resource
resource selection during the activation of the recovery LSP (i.e. selection during the activation of the recovery LSP (i.e. Tas ű Ta).
Tras ű Tra).
D.Papadimitriou et al. - Internet Draft - Expires November 2003 20
5.5.4 LSP Segment Restoration 5.5.4 LSP Segment Restoration
The above approaches can be applied on a sub-network basis rather The above approaches can be applied on an edge-to-edge LSP basis
than end-to-end basis (in order to reduce the global recovery time). rather than end-to-end LSP basis (i.e. to reduce the global recovery
time) by allowing the recovery of the individual LSP segments
constituting the end-to-end LSP.
It should be also noted that using the horizontal hierarchical It should be also noted that using the horizontal hierarchical
approach described in Section 7.1, that a given end-to-end LSP can approach described in Section 7.1, that a given end-to-end LSP can
be recovered by multiple recovery mechanisms (e.g. 1:1 protection in be recovered by multiple recovery mechanisms applied on a segment
a metro edge network but M:N protection in the core). These basis (e.g. 1:1 edge-to-edge LSP protection in a metro network and
mechanisms are ideally independent and may even use different M:N edge-to-edge protection in the core). These mechanisms are
failure localization and notification mechanisms. ideally independent and may even use different failure localization
and notification mechanisms.
6. Normalization 6. Normalization
Normalization is defined as the mechanism allowing switching normal Normalization is defined as the mechanism allowing switching normal
traffic from the recovery LSP/span to the working LSP/span traffic from the recovery LSP/span to the working LSP/span
previously under failure condition. previously under failure condition.
Use of normalization is under the discretion of the recovery domain Use of normalization is under the discretion of the recovery domain
policy. Normalization (reversion) may impact the normal traffic (a policy. Normalization (also referred to as reversion) may impact the
second hit) depending on the normalization mechanism used. normal traffic (a second hit) depending on the normalization
mechanism used.
If normalization is supported 1) the LSP/span must be returned to If normalization is supported 1) the LSP/span must be returned to
the working LSP/span when the failure condition clears 2) capability the working LSP/span when the failure condition clears 2) capability
to de-activate (turn-off) the use of reversion should be provided. to de-activate (turn-off) the use of reversion should be provided.
De-activation of reversion should not impact the normal traffic De-activation of reversion should not impact the normal traffic
(regardless if currently using the working or recovery LSP/span). regardless if currently using the working or recovery LSP/span.
Note: during the failure, the reuse of any non-failed resources Note: during the failure, the reuse of any non-failed resources
(e.g. LSP spans) belonging to the working LSP/span is under the (e.g. LSP and/or spans) belonging to the working LSP/span is under
discretion of recovery domain policy. the discretion of recovery domain policy.
6.1 Wait-To-Restore 6.1 Wait-To-Restore
D.Papadimitriou et al. - Internet Draft ű June 2003 20
A specific mechanism (Wait-To-Restore) is used to prevent frequent A specific mechanism (Wait-To-Restore) is used to prevent frequent
recovery switching operation due to an intermittent defect (e.g. BER recovery switching operation due to an intermittent defect (e.g. BER
fluctuating around the SD threshold). fluctuating around the SD threshold).
First, an LSP/span under failure condition must become fault-free, First, an LSP/span under failure condition must become fault-free,
e.g. a BER less than a certain recovery threshold. After the e.g. a BER less than a certain recovery threshold. After the
recovered LSP/span (i.e. the previously working LSP/span) meets this recovered LSP/span (i.e. the previously working LSP/span) meets this
criterion, a fixed period of time shall elapse before normal traffic criterion, a fixed period of time shall elapse before normal traffic
uses the corresponding resources again. This duration called Wait- uses the corresponding resources again. This duration called Wait-
To-Restore (WTR) period or timer is generally of the order of a few To-Restore (WTR) period or timer is generally of the order of a few
minutes (for instance, 5 minutes) and should be capable of being minutes (for instance, 5 minutes) and should be capable of being
set. The WTR timer may be either a fixed period, or provide for set. The WTR timer may be either a fixed period, or provide for
incremental longer periods before retrying. An SF or SD condition on incremental longer periods before retrying. An SF or SD condition on
the previously working LSP/span will override the WTR timer value the previously working LSP/span will override the WTR timer value
(i.e. the WTR cancels and the WTR timer will restart). (i.e. the WTR cancels and the WTR timer will restart).
D.Papadimitriou et al. - Internet Draft - Expires November 2003 21
6.2 Revertive Mode Operation 6.2 Revertive Mode Operation
In revertive mode of operation, when the recovery LSP/span is no In revertive mode of operation, when the recovery LSP/span is no
longer required, i.e. the failed working LSP/span is no longer in SD longer required, i.e. the failed working LSP/span is no longer in SD
or SF condition, a local Wait-to-Restore (WTR) state will be or SF condition, a local Wait-to-Restore (WTR) state will be
activated before switching the normal traffic back to the recovered activated before switching the normal traffic back to the recovered
working LSP/span. working LSP/span.
During the reversion operation, since this state becomes the highest During the reversion operation, since this state becomes the highest
in priority, signalling must maintain the normal traffic on the in priority, signalling must maintain the normal traffic on the
skipping to change at line 1114 skipping to change at line 1160
recovery LSP/span usage by the normal traffic may be preempted if a recovery LSP/span usage by the normal traffic may be preempted if a
higher priority request for this recovery LSP/span is attempted. higher priority request for this recovery LSP/span is attempted.
6.3 Orphans 6.3 Orphans
When a reversion operation is requested normal traffic must be When a reversion operation is requested normal traffic must be
switched from the recovery to the recovered working LSP/span. A switched from the recovery to the recovered working LSP/span. A
particular situation occurs when the previously working LSP/span can particular situation occurs when the previously working LSP/span can
not be recovered such that normal traffic can not be switched back. not be recovered such that normal traffic can not be switched back.
In such a case, the LSP/span under failure condition (also referred In such a case, the LSP/span under failure condition (also referred
to as ˘orphan÷) must be cleared i.e. removed from the pool of to as "orphan") must be cleared i.e. removed from the pool of
resources allocated for normal traffic. Otherwise, potential de- resources allocated for normal traffic. Otherwise, potential de-
synchronization between the control and transport plane resource synchronization between the control and transport plane resource
usage can appear. Depending on the signalling protocol capabilities usage can appear. Depending on the signalling protocol capabilities
and behavior different mechanisms are to be expected here. and behavior different mechanisms are to be expected here.
Therefore any reserved or allocated resources for the LSP/span under Therefore any reserved or allocated resources for the LSP/span under
failure condition must be unreserved/de-allocated. Several ways can failure condition must be unreserved/de-allocated. Several ways can
be used for that purpose: wait for the elapsing of the clear-out be used for that purpose: wait for the elapsing of the clear-out
time interval, or initiate a deletion from the ingress or the egress time interval, or initiate a deletion from the ingress or the egress
node, or trigger the initiation of deletion from an entity (such as node, or trigger the initiation of deletion from an entity (such as
D.Papadimitriou et al. - Internet Draft ű June 2003 21
an EMS or NMS) capable to react on the reception of an appropriate an EMS or NMS) capable to react on the reception of an appropriate
notification message. notification message.
7. Hierarchies 7. Hierarchies
Recovery mechanisms are being made available at multiple (if not Recovery mechanisms are being made available at multiple (if not
each) transport layers within so-called ˘IP-over-optical÷ networks. each) transport layers within so-called "IP/MPLS-over-optical"
However, each layer has certain recovery features and one needs to networks. However, each layer has certain recovery features and one
determine the exact impact of the interaction between the recovery needs to determine the exact impact of the interaction between the
mechanisms provided by these layers. recovery mechanisms provided by these layers.
Hierarchies are used to build scalable complex systems. Abstraction Hierarchies are used to build scalable complex systems. Abstraction
is used as a mechanism to build large networks or as a technique for is used as a mechanism to build large networks or as a technique for
enforcing technology, topological or administrative boundaries. The enforcing technology, topological or administrative boundaries. The
same hierarchical concept can be applied to control the network same hierarchical concept can be applied to control the network
D.Papadimitriou et al. - Internet Draft - Expires November 2003 22
survivability. In general, it is expected that the recovery action survivability. In general, it is expected that the recovery action
is taken by the recoverable LSP/span closest to the failure in order is taken by the recoverable LSP/span closest to the failure in order
to avoid the multiplication of recovery actions. Moreover, recovery to avoid the multiplication of recovery actions. Moreover, recovery
hierarchies can be also bound to control plane logical partitions hierarchies can be also bound to control plane logical partitions
(e.g. administrative or topological boundaries). Each of them may (e.g. administrative or topological boundaries). Each of them may
apply different recovery mechanisms. apply different recovery mechanisms.
In brief, commonly accepted ideas are generally that the lower In brief, commonly accepted ideas are generally that the lower
layers can provide coarse but faster recovery while the higher layers can provide coarse but faster recovery while the higher
layers can provide finer but slower recovery. Moreover, it is also layers can provide finer but slower recovery. Moreover, it is also
more than desirable to avoid too many layers with functional desirable to avoid that similar layers with functional overlaps to
overlaps. In this context, this section intends to analyze these optimize network resource utilization and processing overhead. In
hierarchical aspects including the physical (passive) layer(s). this context, this section intends to analyze these hierarchical
aspects including the physical (passive) layer(s).
7.1 Horizontal Hierarchy (Partitioning) 7.1 Horizontal Hierarchy (Partitioning)
A horizontal hierarchy is defined when partitioning a single layer A horizontal hierarchy is defined when partitioning a single layer
network (and its control plane) into several recovery domains. network (and its control plane) into several recovery domains.
Within a domain, the recovery scope may extend over a link (or Within a domain, the recovery scope may extend over a link (or
span), LSP segment or even an end-to-end LSP. Moreover, an span), LSP segment or even an end-to-end LSP. Moreover, an
administrative domain may consist of a single recovery domain or can administrative domain may consist of a single recovery domain or can
be partitioned into several smaller recovery domains. The operator be partitioned into several smaller recovery domains. The operator
can partition the network into recovery domains based on physical can partition the network into recovery domains based on physical
network topology, control plane capabilities or various traffic network topology, control plane capabilities or various traffic
engineering constraints. engineering constraints.
An example often addressed in the literature is the metro-core-metro An example often addressed in the literature is the metro-core-metro
application (sometimes extended to a metro-metro/core-core) within a application (sometimes extended to a metro-metro/core-core) within a
single transport layer (see Section 7.2). For such a case, an end- single transport layer (see Section 7.2). For such a case, an end-
to-end LSP is defined between the ingress and egress metro nodes, to-end LSP is defined between the ingress and egress metro nodes,
while LSP segments may be defined within the metro or core sub- while LSP segments may be defined within the metro or core sub-
networks. Each of these topological structures determines a so- networks. Each of these topological structures determines a so-
called ˘recovery domain÷ since each of the LSPs they carry can have called "recovery domain" since each of the LSPs they carry can have
its own recovery type (or even scheme). The support of multiple its own recovery type (or even scheme). The support of multiple
recovery schemes within a sub-network is referred to as a multi- recovery types and schemes within a sub-network is referred to as a
recovery capable domain or simply multi-recovery domain. multi-recovery capable domain or simply multi-recovery domain.
7.2 Vertical Hierarchy (Layers) 7.2 Vertical Hierarchy (Layers)
D.Papadimitriou et al. - Internet Draft ű June 2003 22
It is a very challenging task to combine in a coordinated manner the It is a very challenging task to combine in a coordinated manner the
different recovery capabilities available across the path (i.e. different recovery capabilities available across the path (i.e.
switching capable) and section layers to ensure that certain network switching capable) and section layers to ensure that certain network
survivability objectives are met for the different services survivability objectives are met for the different services
supported by the network. supported by the network.
As a first analysis step, one can draw the following guidelines for As a first analysis step, one can draw the following guidelines for
a vertical coordination of the recovery mechanisms: a vertical coordination of the recovery mechanisms:
- The lower the layer the faster the notification and switching - The lower the layer the faster the notification and switching
- The higher the layer the finer the granularity of the recoverable - The higher the layer the finer the granularity of the recoverable
entity and therefore the granularity of the recovery resource entity and therefore the granularity of the recovery resource
(and subsequently its sharing ratio) (and subsequently its sharing ratio)
D.Papadimitriou et al. - Internet Draft - Expires November 2003 23
Therefore, in the scope of this analysis, a vertical hierarchy Therefore, in the scope of this analysis, a vertical hierarchy
consists of multiple layered transport planes providing different: consists of multiple layered transport planes providing different:
- Discrete bandwidth granularities for non-packet LSPs such as OCh, - Discrete bandwidth granularities for non-packet LSPs such as OCh,
ODUk, HOVC/STS-SPE and LOVC/VT-SPE LSPs and continuous bandwidth ODUk, STS_SPE/HOVC and VT_SPE/LOVC LSPs and continuous bandwidth
granularities for packet LSPs granularities for packet LSPs
- Potentially, recovery capabilities with different temporal - Potentially, recovery capabilities with different temporal
granularities: ranging from milliseconds to tens of seconds granularities: ranging from milliseconds to tens of seconds
Note: based on the bandwidth granularity we can determine four Note: based on the bandwidth granularity we can determine four
classes of vertical hierarchiesĂ (1) packet over packet (2) packet classes of vertical hierarchies (1) packet over packet (2) packet
over circuit (3) circuit over packet and (4) circuit over circuit. over circuit (3) circuit over packet and (4) circuit over circuit.
Here below we extend a little bit more on (4), (2) being covered in Here below we extend a little bit more on (4), (2) being covered in
[TE-RH] on the other hand (1) is extensively covered at the MPLS [RFC 3386]. On the other hand (1) is extensively covered at the MPLS
Working Group, and (3) at the PWE3 Working Group. Working Group, and (3) at the PWE3 Working Group.
In SDH/Sonet environments, one typically considers the LOVC/VT and In SDH/Sonet environments, one typically considers the VT_SPE/LOVC
HOVC/STS SPE as independent layers, LOVC/VT LSP using the underlying and STS SPE/HOVC as independent layers, VT_SPE/LOVC LSP using the
HOVC/STS SPE LSPs as links, for instance. In OTN, the ODUk path underlying STS_SPE/HOVC LSPs as links, for instance. In OTN, the
layers will lie on the OCh path layer i.e. the ODUk LSPs using the ODUk path layers will lie on the OCh path layer i.e. the ODUk LSPs
underlying OCh LSPs as links. Notice here that server layer LSPs may using the underlying OCh LSPs as OTUk links. Note here that lower
simply be provisioned and not dynamically triggered or established layer LSPs may simply be provisioned and not necessarily dynamically
(control driven approach). triggered or established (control driven approach). In this context,
an LSP at the path layer (i.e. established using GMPLS signalling),
The following figure (including only the path layers) illustrates for instance an optical channel LSP, appears at the OTUk layer as a
the hierarchy that can be covered by the recovery architecture of a link, typically controlled by a link management protocol such as
network comprising a SDH/Sonet and an OTN part: LMP.
LOVC <------------------------------------------------------> LOVC
| |
HOVC ---- HOVC <----------------------------------> HOVC ---- HOVC
| |
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
| |
ODUk ---- ODUk <--------------> ODUk ---- ODUk
| |
OTUk <--------------> OTUk
| |
OCh -- OCh -..- OCh -- OCh
D.Papadimitriou et al. - Internet Draft ű June 2003 23
In this context, the important points are the following:
- these layers are path layers; i.e. the ones controlled by
the GMPLS (in particular, signalling) protocol suite.
- an LSP at the lower layer for instance an optical channel (=
network connection) appears as a section (= link) for the OTUk
layer i.e. the links that are typically controlled by link
management protocols such as LMP.
The first key issue with multi-layer recovery is that achieving The first key issue with multi-layer recovery is that achieving
control plane individual or bulk LSP recovery will be as efficient control plane individual or bulk LSP recovery will be as efficient
as the underlying link (local span) recovery. In such a case, the as the underlying link (local span) recovery. In such a case, the
span can be either protected or unprotected, but the LSP it carries span can be either protected or unprotected, but the LSP it carries
MUST be (at least locally) recoverable. Therefore, the span recovery MUST be (at least locally) recoverable. Therefore, the span recovery
process can either be independent when protected (or restorable), or process can either be independent when protected (or restorable), or
triggered by the upper LSP recovery process. The former requires triggered by the upper LSP recovery process. The former requires
coordination in order to achieve subsequent LSP recovery. Therefore, coordination in order to achieve subsequent LSP recovery. Therefore,
in order to achieve robustness and fast convergence, multi-layer in order to achieve robustness and fast convergence, multi-layer
skipping to change at line 1268 skipping to change at line 1294
simultaneous recovery actions that may lead to race conditions and simultaneous recovery actions that may lead to race conditions and
in turn, reduce the optimization of the resource utilization and/or in turn, reduce the optimization of the resource utilization and/or
generate global instabilities in the network (see [MANCHESTER]). generate global instabilities in the network (see [MANCHESTER]).
Therefore, a consistent and efficient escalation strategy is needed Therefore, a consistent and efficient escalation strategy is needed
to coordinate recovery across several layers. to coordinate recovery across several layers.
Therefore, one can expect that the definition of the recovery Therefore, one can expect that the definition of the recovery
mechanisms and protocol(s) is technology independent such that they mechanisms and protocol(s) is technology independent such that they
can be consistently implemented at different layers; this would in can be consistently implemented at different layers; this would in
turn simplify their global coordination. Moreover, as mentioned in turn simplify their global coordination. Moreover, as mentioned in
[TE-RH], some looser form of coordination and communication between [RFC 3386], some looser form of coordination and communication
(vertical) layers such a consistent hold-off timer configuration
(and setup through signalling during the working LSP establishment) D.Papadimitriou et al. - Internet Draft - Expires November 2003 24
can be considered in this context, allowing synchronization between between (vertical) layers such a consistent hold-off timer
recovery actions performed across these layers. configuration (and setup through signalling during the working LSP
establishment) can be considered in this context, allowing
synchronization between recovery actions performed across these
layers.
Note: Recovery Granularity Note: Recovery Granularity
In most environments, the design of the network and the vertical In most environments, the design of the network and the vertical
distribution of the LSP bandwidth are such that the recovery distribution of the LSP bandwidth are such that the recovery
granularity is finer for higher layers. The OTN and SDH/Sonet layers granularity is finer for higher layers. The OTN and Sonet/SDH layers
can only recover the whole section or the individual connections it can only recover the whole section or the individual connections it
transports whereas IP/MPLS layer(s) can recover individual packet transports whereas IP/MPLS layer(s) can recover individual packet
LSPs or groups of packet LSPs. LSPs or groups of packet LSPs.
Obviously, the recovery granularity at the sub-wavelength (i.e. Obviously, the recovery granularity at the sub-wavelength (i.e.
SDH/Sonet) level can be provided only when the network includes Sonet/SDH) level can be provided only when the network includes
devices switching at the same granularity level (and thus not with devices switching at the same granularity level (and thus not with
optical channel switching capable devices). Therefore, the network optical channel switching capable devices). Therefore, the network
layer can deliver control-plane driven recovery mechanisms on a per- layer can deliver control-plane driven recovery mechanisms on a per-
D.Papadimitriou et al. - Internet Draft ű June 2003 24
LSP basis if and only if the LSPs class has the corresponding LSP basis if and only if the LSPs class has the corresponding
switching capability at the transport plane level. switching capability at the transport plane level.
7.3 Escalation Strategies 7.3 Escalation Strategies
There are two types of escalation strategies (see [DEMEESTER]): There are two types of escalation strategies (see [DEMEESTER]):
bottom-up and top-down. bottom-up and top-down.
The bottom-up approach assumes that lower layer recovery schemes are The bottom-up approach assumes that lower layer recovery types and
more expedient and faster than the upper layer one. Therefore we can schemes are more expedient and faster than the upper layer one.
inhibit or hold-off higher layer recovery. However this assumption Therefore we can inhibit or hold-off higher layer recovery. However
is not entirely true. Imagine a SDH/Sonet based protection mechanism this assumption is not entirely true. Consider a Sonet/SDH based
(with a less than 50 ms protection switching time) lying on top of protection mechanism (with a less than 50 ms protection switching
an OTN restoration mechanism (with a less than 200 ms restoration time) lying on top of an OTN restoration mechanism (with a less than
time). Therefore, this assumption should be (at least) clarified as: 200 ms restoration time). Therefore, this assumption should be (at
lower layer recovery schemes are faster than upper level one but least) clarified as: lower layer recovery types and schemes are
only if the same type of recovery mechanism is used at each layer faster than upper level one but only if the same type of recovery
(assuming that the lower layer one is faster). mechanism is used at each layer (assuming that the lower layer one
is faster).
Consequently, taking into account the recovery actions at the Consequently, taking into account the recovery actions at the
different layers in a bottom-up approach, if lower layer recovery different layers in a bottom-up approach, if lower layer recovery
mechanisms are provided and sequentially activated in conjunction mechanisms are provided and sequentially activated in conjunction
with higher layer ones, the lower layers MUST have an opportunity to with higher layer ones, the lower layers MUST have an opportunity to
recover normal traffic before the higher layers do. However, if recover normal traffic before the higher layers do. However, if
lower layer recovery is slower than higher layer recovery, the lower lower layer recovery is slower than higher layer recovery, the lower
layer MUST either communicate the failure related information to the layer MUST either communicate the failure related information to the
higher layer(s) (and allow it to perform recovery), or use a hold- higher layer(s) (and allow it to perform recovery), or use a hold-
off timer in order to temporarily set the higher layer recovery off timer in order to temporarily set the higher layer recovery
action in a ˘standby mode÷. Note that the a priori information action in a "standby mode". Note that the a priori information
exchange between layers concerning their efficiency is not within exchange between layers concerning their efficiency is not within
the current scope of this document. Nevertheless, the coordination the current scope of this document. Nevertheless, the coordination
functionality between layers must be configurable and tunable. functionality between layers must be configurable and tunable.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 25
An example of coordination between the optical and packet layer An example of coordination between the optical and packet layer
control plane enables for instance letting the optical layer control plane enables for instance letting the optical layer
performing the failure management operations (in particular, failure performing the failure management operations (in particular, failure
detection and notification) while giving to the packet layer control detection and notification) while giving to the packet layer control
plane the authority to perform the recovery actions. In case of plane the authority to perform the recovery actions. In case of
packet layer unsuccessful recovery action, fallback at the optical packet layer unsuccessful recovery action, fallback at the optical
layer can be subsequently performed. layer can be subsequently performed.
The top-down approach attempts service recovery at the higher layers The Top-down approach attempts service recovery at the higher layers
before invoking lower layer recovery. Higher layer recovery is before invoking lower layer recovery. Higher layer recovery is
service selective, and permits "per-CoS" or "per-connection" re- service selective, and permits "per-CoS" or "per-connection" re-
routing. With this approach, the most important aspect is that the routing. With this approach, the most important aspect is that the
upper layer must provide its own reliable and independent failure upper layer must provide its own reliable and independent failure
detection mechanism from the lower layer. detection mechanism from the lower layer.
The same reference suggests also recovery mechanisms incorporating a The same reference suggests also recovery mechanisms incorporating a
coordinated effort shared by two adjacent layers with periodic coordinated effort shared by two adjacent layers with periodic
status updates. Moreover, at certain layers, some of these recovery status updates. Moreover, at certain layers, some of these recovery
operations can be pre-assigned, e.g. a particular link will be operations can be pre-assigned, e.g. a particular link will be
D.Papadimitriou et al. - Internet Draft ű June 2003 25
handled by the packet layer while another will be handled by the handled by the packet layer while another will be handled by the
optical layer. optical layer.
7.4 Disjointness 7.4 Disjointness
Having link and node diverse working and recovery LSPs/spans does Having link and node diverse working and recovery LSPs/spans does
not guarantee working and recovery LSPs/Spans disjointness. Due to not guarantee working and recovery LSPs/Spans disjointness. Due to
the common physical layer topology (passive), additional the common physical layer topology (passive), additional
hierarchical concepts such as the Shared Risk Link Group (SRLG) and hierarchical concepts such as the Shared Risk Link Group (SRLG) and
mechanisms such as SRLG diverse path computation must be developed mechanisms such as SRLG diverse path computation must be developed
skipping to change at line 1377 skipping to change at line 1404
The SRLG properties can be summarized as follows: The SRLG properties can be summarized as follows:
1) A link belongs to more than one SRLG if and only if it crosses 1) A link belongs to more than one SRLG if and only if it crosses
one of the resources covered by each of them. one of the resources covered by each of them.
2) Two links belonging to the same SRLG can belong individually to 2) Two links belonging to the same SRLG can belong individually to
(one or more) other SRLGs. (one or more) other SRLGs.
3) The SRLG set S of an LSP is defined as the union of the 3) The SRLG set S of an LSP is defined as the union of the
D.Papadimitriou et al. - Internet Draft - Expires November 2003 26
individual SRLG s of the individual links composing this LSP. individual SRLG s of the individual links composing this LSP.
SRLG disjointness for LSP: SRLG disjointness for LSP:
The LSP SRLG disjointness concept is based on the following The LSP SRLG disjointness concept is based on the following
postulate: an LSP (i.e. sequence of links and nodes) covers an postulate: an LSP (i.e. sequence of links and nodes) covers an
SRLG if and only if it crosses one of the links or nodes SRLG if and only if it crosses one of the links or nodes
belonging to that SRLG. belonging to that SRLG.
Therefore, the SRLG disjointness for LSPs can be defined as Therefore, the SRLG disjointness for LSPs can be defined as
follows: two LSPs are disjoint with respect to an SRLG s if and follows: two LSPs are disjoint with respect to an SRLG s if and
only if none of them covers simultaneously this SRLG. only if they do not cover simultaneously this SRLG s.
Whilst the SRLG disjointness for LSPs with respect of a set S of Whilst the SRLG disjointness for LSPs with respect to a set S of
SRLGs is defined as follows: two LSPs are disjoint with respect SRLGs is defined as follows: two LSPs are disjoint with respect
to a set of SRLGs S if and only if the set of SRLGs they to a set of SRLGs S if and only if the common SRLGs between the
individually cover are mutually disjoint from the set S. sets of SRLGs they individually cover is disjoint from set S.
The impact on recovery is obvious: SRLG disjointness is a necessary The impact on recovery is obvious: SRLG disjointness is a necessary
(but not a sufficient) condition to ensure optical network (but not a sufficient) condition to ensure optical network
D.Papadimitriou et al. - Internet Draft ű June 2003 26
survivability. With respect to the physical network resources, a survivability. With respect to the physical network resources, a
working-recovery LSP/span pair must be SRLG disjoint in case of working-recovery LSP/span pair must be SRLG disjoint in case of
dedicated recovery type while a working-recovery LSP/span group must dedicated recovery type. On the other hand, in case of shared
be SRLG disjoint in case of shared recovery. recovery, a group of working LSP/span must be mutually SRLG disjoint
in order to allow for a (single and common) shared recovery LSP
itself SRLG disjoint from each of the working LSP/span.
8. Recovery Scheme/Strategy Selection 8. Recovery Type/Scheme Analysis
In order to provide a structured selection and analysis of the In order to provide a structured analysis of the recovery types and
recovery scheme/strategy, the following dimensions can be defined: schemes, the following dimensions can be considered:
1. Fast convergence (performance): provide a mechanism that 1. Fast convergence (performance): provide a mechanism that
aggregates multiple failures (this implies fast failure aggregates multiple failures (this implies fast failure
detection and correlation mechanisms) and fast recovery decision detection and correlation mechanisms) and fast recovery decision
independently of the number of failures occurring in the optical independently of the number of failures occurring in the optical
network (implying also a fast failure notification). network (implying also a fast failure notification).
2. Efficiency (scalability): minimize the switching time required 2. Efficiency (scalability): minimize the switching time required
for LSP/span recovery independently of number of LSPs/spans being for LSP/span recovery independently of number of LSPs/spans being
recovered (this implies an efficient failure correlation, a fast recovered (this implies an efficient failure correlation, a fast
skipping to change at line 1430 skipping to change at line 1459
3. Robustness (availability): minimize the LSP/span downtime 3. Robustness (availability): minimize the LSP/span downtime
independently of the underlying topology of the transport plane independently of the underlying topology of the transport plane
(this implies a highly responsive recovery mechanism). (this implies a highly responsive recovery mechanism).
4. Resource optimization (optimality): minimize the resource 4. Resource optimization (optimality): minimize the resource
capacity, including LSP/span and nodes (switching capacity), capacity, including LSP/span and nodes (switching capacity),
required for recovery purposes; this dimension can also be required for recovery purposes; this dimension can also be
referred to as optimize the sharing degree of the recovery referred to as optimize the sharing degree of the recovery
resources. resources.
5. Cost optimization: provide a cost-effective recovery strategy. D.Papadimitriou et al. - Internet Draft - Expires November 2003 27
5. Cost optimization: provide a cost-effective recovery type/scheme.
However, these dimensions are either out of the scope of this However, these dimensions are either out of the scope of this
document such as cost optimization and recovery path computational document such as cost optimization and recovery path computational
aspects or going in opposite directions. For instance, it is obvious aspects or going in opposite directions. For instance, it is obvious
that providing a 1+1 recovery type for each LSP minimizes the LSP that providing a 1+1 LSP recovery type minimizes the LSP downtime
downtime (in case of failure) while being non-scalable and recovery (in case of failure) while being non-scalable and recovery resource
resource consuming without enabling any extra-traffic. consuming without enabling any extra-traffic.
The following sections try to provide a first response in order to The following sections provide an analysis of the recovery types
select a recovery strategy with respect to the dimensions described (and schemes) proposed in [CCAMP-TERM] with respect to the
above and the recovery schemes proposed in [CCAMP-TERM]. dimensions described above and assess the current GMPLS
capabilities. In turn, this allows evaluating the need for further
GMPLS signalling or routing extensions.
8.1 Fast Convergence (Detection/Correlation and Hold-off Time) 8.1 Fast Convergence (Detection/Correlation and Hold-off Time)
Fast convergence is related to the failure management operations. It Fast convergence is related to the failure management operations. It
refers to the elapsing time between the failure detection/ refers to the elapsing time between the failure detection/
correlation and hold-off time, point at which the recovery switching correlation and hold-off time, point at which the recovery switching
actions are initiated. This point has been already discussed in actions are initiated. This point has been already discussed in
Section 4. Section 4.
8.2 Efficiency (Switching Time) 8.2 Efficiency (Switching Time)
D.Papadimitriou et al. - Internet Draft ű June 2003 27
In general, the more pre-assignment/pre-planning of the recovery In general, the more pre-assignment/pre-planning of the recovery
LSP/span, the more rapid the recovery scheme is. Since protection LSP/span, the more rapid the recovery is. Since protection implies
implies pre-assignment (and cross-connection in case of LSP pre-assignment (and cross-connection) of the protection resources,
recovery) of the protection resources, in general, protection in general, protection recover faster than restoration.
schemes recover faster than restoration schemes.
Span restoration (since using control plane) is also likely to be Span restoration (since using control plane) is also likely to be
slower than most span protection types; however this greatly depends slower than most span protection types; however this greatly depends
on the span restoration signalling efficiency. LSP Restoration with on the span restoration signalling efficiency. LSP Restoration with
pre-signaled and pre-selected recovery resources is likely to be pre-signaled and pre-selected recovery resources is likely to be
faster than fully dynamic LSP restoration, especially because of the faster than fully dynamic LSP restoration, especially because of the
elimination of any potential crank-back during the recovery LSP elimination of any potential crank-back during the recovery LSP
establishment. establishment.
If one excludes the crank-back issue, the difference between dynamic If one excludes the crank-back issue, the difference between dynamic
and pre-planned restoration depends on the restoration path and pre-planned restoration depends on the restoration path
computation and path selection time. Since computational computation and selection time. Since computational considerations
considerations are outside of the scope of this document, it is up are outside of the scope of this document, it is up to the vendor to
to the vendor to determine the average path computation time in determine the average path computation time in different scenarios
different scenarios and to the operator to decide whether or not and to the operator to decide whether or not dynamic restoration is
dynamic restoration is advantageous over pre-planned schemes advantageous over pre-planned schemes depending on the network
depending on the network environment. This difference depends also environment. This difference depends also on the flexibility
on the flexibility provided by pre-planned restoration with respect provided by pre-planned restoration with respect to dynamic one: the
to dynamic one: the former implies a limited number of failure former implies a limited number of failure scenarios (that can be
scenarios (that can be due for instance to local storage due for instance to local storage limitation). This, while the
limitation). This, while the latter enables an on-demand path latter enables an on-demand path computation based on the
computation based on the information received through failure information received through failure notification and as such more
notification and as such more robust with respect to the failure robust with respect to the failure scenario scope.
scenario scope.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 28
Moreover, LSP segment restoration, in particular, dynamic Moreover, LSP segment restoration, in particular, dynamic
restoration (i.e. no path pre-computation so none of the recovery restoration (i.e. no path pre-computation so none of the recovery
resource is pre-signaled) will generally be faster than end-to-end resource is pre-signaled) will generally be faster than an end-to-
LSP schemes. However, local LSP restoration assumes that each LSP end LSP recovery. However, local LSP restoration assumes that each
segment end-point has enough computational capacity to perform this LSP segment end-point has enough computational capacity to perform
operation while end-to-end requires only that LSP end-points this operation while end-to-end requires only that LSP end-points
provides this path computation capability. provides this path computation capability.
Recovery time objectives for SDH/Sonet protection switching (not Recovery time objectives for Sonet/SDH protection switching (not
including time to detect failure) are specified in [G.841] at 50 ms, including time to detect failure) are specified in [G.841] at 50 ms,
taking into account constraints on distance, number of connections taking into account constraints on distance, number of connections
involved, and in the case of ring enhanced protection, number of involved, and in the case of ring enhanced protection, number of
nodes in the ring. Recovery time objectives for restoration nodes in the ring. Recovery time objectives for restoration
mechanisms have been proposed through a separate effort [TE-RH]. mechanisms have been proposed through a separate effort [RFC 3386].
8.3 Robustness 8.3 Robustness
In general, the less pre-assignment (protection)/pre-planning In general, the less pre-assignment (protection)/pre-planning
(restoration) of the recovery LSP/span, the more robust the recovery (restoration) of the recovery LSP/span, the more robust the recovery
type/scheme is to a variety of (single) failures, provided that type or scheme is to a variety of single failures, provided that
adequate resources are available. Moreover, the pre-selection of the adequate resources are available. Moreover, the pre-selection of the
recovery resources gives less flexibility for multiple failure recovery resources gives less flexibility for multiple failure
D.Papadimitriou et al. - Internet Draft ű June 2003 28
scenarios than no recovery resource pre-selection. For instance, if scenarios than no recovery resource pre-selection. For instance, if
failures occur that affect two LSPs sharing a common link along failures occur that affect two LSPs sharing a common link along
their restoration paths, then only one of these LSPs can be their restoration paths, then only one of these LSPs can be
recovered. This occurs unless the restoration path of at least one recovered. This occurs unless the restoration path of at least one
of these LSPs is re-computed or the local resource assignment is of these LSPs is re-computed or the local resource assignment is
modified on the fly. modified on the fly.
In addition, recovery schemes with pre-planned recovery resources, In addition, recovery types and schemes with pre-planned recovery
in particular spans for protection and LSP for restoration purposes, resources, in particular LSP/spans for protection and LSP for
will not be able to recover from failures that simultaneously affect restoration purposes, will not be able to recover from failures that
both the working and recovery LSP/span. Thus, the recovery resources simultaneously affect both the working and recovery LSP/span. Thus,
should ideally be chosen to be as disjoint as possible (with respect the recovery resources should ideally be as disjoint as possible
to link, node and SRLG) from the working ones, so that any single (with respect to link, node and SRLG) from the working ones, so that
failure event will not affect both working and recovery LSP/span. In any single failure event will not affect both working and recovery
brief, working and recovery resource must be fully diverse in order LSP/span. In brief, working and recovery resource must be fully
to guarantee that a given failure will not affect simultaneously the diverse in order to guarantee that a given failure will not affect
working and the recovery LSP/span. Also, the risk of simultaneous simultaneously the working and the recovery LSP/span. Also, the risk
failure of the working and restoration LSP can be reduced by re- of simultaneous failure of the working and recovery LSP can be
computing a restoration path whenever a failure occurs along the reduced by computing a new recovery path whenever a failure occurs
corresponding recovery LSP or by re-computing a restoration path and along one of the recovery LSPs or by computing a new recovery path
re-provisioning the corresponding recovery LSP whenever a failure and provision the corresponding LSP whenever a failure occurs along
occurs along a working LSP/span. This method enables to maintain the a working LSP/span. Both methods enable to maintain the number of
number of available recovery path constant. available recovery path constant.
The robustness of a recovery scheme is also determined by the amount The robustness of a recovery scheme is also determined by the amount
of reserved (i.e. signaled) recovery resources within a given shared of reserved (i.e. signaled) recovery resources within a given shared
resource pool: as the amount of recovery resources sharing degree resource pool: as the amount of recovery resources sharing degree
increases, the recovery scheme becomes less robust to multiple increases, the recovery scheme becomes less robust to multiple
failure occurrences. Recovery schemes, in particular restoration, LSP/span failure occurrences. Recovery schemes, in particular
with pre-signaled resource reservation (with or without pre- restoration, with pre-signaled resource reservation (with or without
selection) should be capable to reserve the adequate amount of pre-selection) should be capable to reserve the adequate amount of
D.Papadimitriou et al. - Internet Draft - Expires November 2003 29
resource to ensure recovery from any specific set of failure events, resource to ensure recovery from any specific set of failure events,
such as any single SRLG failure, any two SRLG failures etc. such as any single SRLG failure, any two SRLG failures etc.
8.4 Resource Optimization 8.4 Resource Optimization
It is commonly admitted that sharing recovery resources provides It is commonly admitted that sharing recovery resources provides
network resource optimization. Therefore, from a resource network resource optimization. Therefore, from a resource
utilization perspective, protection schemes are often classified utilization perspective, protection schemes are often classified
with respect to their degree of sharing recovery resources with with respect to their degree of sharing recovery resources with
respect to the working entities. Moreover, non-permanent bridging respect to the working entities. Moreover, non-permanent bridging
protection types allow (under normal conditions) for extra-traffic protection types allow (under normal conditions) for extra-traffic
over the recovery resources. over the recovery resources.
From this perspective 1) 1+1 LSP/Span protection is the more From this perspective 1) 1+1 LSP/Span protection is the more
resource consuming protection type since it doesnĂt allow for any resource consuming protection type since it doesn't allow for any
extra-traffic 2) 1:1 LSP/span protection type requires dedicated extra-traffic 2) 1:1 LSP/span protection type requires dedicated
recovery LSP/span allowing carrying extra preemptible traffic 3) 1:N recovery LSP/span allowing carrying extra preemptible traffic 3) 1:N
and M:N LSP/span recovery types require 1 (or M, respectively) and M:N LSP/span recovery types require 1 (or M, respectively)
recovery LSP/span (shared between the N working LSP/span) while recovery LSP/span (shared between the N working LSP/span) while
allowing carrying extra preemptible traffic. Obviously, 1+1 allowing carrying extra preemptible traffic. Obviously, 1+1
protection precludes and 1:1 recovery type does not allow for protection precludes and 1:1 recovery type does not allow for
recovery LSP/span sharing whereas 1:N and M:N recovery types do recovery LSP/span sharing whereas 1:N and M:N recovery types do
D.Papadimitriou et al. - Internet Draft ű June 2003 29
allow sharing of 1 (M, respectively) recovery LSP/spans between N allow sharing of 1 (M, respectively) recovery LSP/spans between N
working LSP/spans. working LSP/spans.
However, despite the fact that the 1:1 recovery type does not allow However, despite the fact that the 1:1 recovery type does not allow
recovery LSP/span sharing, the recovery schemes (see Section 5.4) recovery LSP/span sharing, the recovery schemes (see Section 5.4)
that can be built from them (e.g.(1:1)^n) do allow for sharing of that can be built from them (e.g.(1:1)^n) do allow for sharing of
recovery resources these entities includes. In addition, the recovery resources these entities includes. In addition, the
flexibility in the usage of shared recovery resources (in flexibility in the usage of shared recovery resources (in
particular, shared links) may be limited because of network topology particular, shared links) may be limited because of network topology
restrictions, e.g. fixed ring topology for traditional enhanced restrictions, e.g. fixed ring topology for traditional enhanced
skipping to change at line 1595 skipping to change at line 1623
(thus over preemptible LSP/span) when the corresponding resources (thus over preemptible LSP/span) when the corresponding resources
have not been committed for LSP/span recovery purposes. have not been committed for LSP/span recovery purposes.
From this, it clearly follows that less recovery resources (i.e. From this, it clearly follows that less recovery resources (i.e.
LSP/spans and switching capacity) have to be allocated to a shared LSP/spans and switching capacity) have to be allocated to a shared
recovery resource pool if a greater sharing degree is allowed. Thus, recovery resource pool if a greater sharing degree is allowed. Thus,
the degree to which the network is survivable is determined by the the degree to which the network is survivable is determined by the
policy that defines the amount of reserved (shared) recovery policy that defines the amount of reserved (shared) recovery
resources and the maximum sharing degree allowed. resources and the maximum sharing degree allowed.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 30
8.4.1. Recovery Resource Sharing 8.4.1. Recovery Resource Sharing
When recovery resources are shared over several LSP/Spans, [GMPLS- When recovery resources are shared over several LSP/Spans, [GMPLS-
RTG], the use of the Maximum Reservable Bandwidth, the Unreserved RTG], the use of the Maximum Reservable Bandwidth, the Unreserved
Bandwidth and the Maximum LSP Bandwidth Link sub-TLVs provides the Bandwidth and the Maximum LSP Bandwidth Link sub-TLVs provides the
information needed to obtain the optimization of the network information needed to obtain the optimization of the network
resources allocated for shared recovery purposes. resources allocated for shared recovery purposes.
The Maximum Reservable Bandwidth is defined as the maximum link The Maximum Reservable Bandwidth is defined as the maximum link
capacity but may be greater in case of link over-subscription. The capacity but may be greater in case of link over-subscription. The
skipping to change at line 1616 skipping to change at line 1646
yet reserved on a given TE link (initial value at each priority yet reserved on a given TE link (initial value at each priority
level corresponds to the Maximum Reservable Bandwidth). Last, the level corresponds to the Maximum Reservable Bandwidth). Last, the
Maximum LSP Bandwidth (per priority) is defined as the smaller of Maximum LSP Bandwidth (per priority) is defined as the smaller of
Unreserved Bandwidth and Maximum Reservable Bandwidth. Unreserved Bandwidth and Maximum Reservable Bandwidth.
Here, one generally considers a recovery resource sharing ratio (or Here, one generally considers a recovery resource sharing ratio (or
degree) in order to globally optimize the shared recovery resource degree) in order to globally optimize the shared recovery resource
usage. The distribution of the bandwidth utilization per (bundled) usage. The distribution of the bandwidth utilization per (bundled)
TE link can be inferred from the per-priority bandwidth pre- TE link can be inferred from the per-priority bandwidth pre-
allocation. This by using the Maximum LSP Bandwidth and the allocation. This by using the Maximum LSP Bandwidth and the
D.Papadimitriou et al. - Internet Draft ű June 2003 30
Unreserved Bandwidth (see [GMPLS-RTG]), the amount of resources Unreserved Bandwidth (see [GMPLS-RTG]), the amount of resources
(over-provisioned) for shared recovery purposes is known from the (over-provisioned) for shared recovery purposes is known from the
IGP. IGP.
In order to analyze this behavior, we define the difference between In order to analyze this behavior, we define the difference between
the Maximum Reservable Bandwidth (in the present case, this value is the Maximum Reservable Bandwidth (in the present case, this value is
greater than the maximum link capacity) and the Maximum LSP greater than the maximum link capacity) and the Maximum LSP
Bandwidth (in the present case, this value corresponds to the Bandwidth (in the present case, this value corresponds to the
Unreserved Bandwidth) per TE link i as the Maximum Sharable Unreserved Bandwidth) per TE link i as the Maximum Sharable
Bandwidth or max_R[i]. Within this quantity, the amount of bandwidth Bandwidth or max_R[i]. Within this quantity, the amount of bandwidth
skipping to change at line 1645 skipping to change at line 1673
optimize the usage of the resources allocated (per TE link) for optimize the usage of the resources allocated (per TE link) for
shared recovery. If one refers to r[i] as the actual bandwidth per shared recovery. If one refers to r[i] as the actual bandwidth per
TE link i (in terms of per component bandwidth unit) committed for TE link i (in terms of per component bandwidth unit) committed for
shared recovery, then the following quantity must be maximized over shared recovery, then the following quantity must be maximized over
the potential TE link candidates: sum {i=1}^N [(R{i} - r{i})/(t{i} ű the potential TE link candidates: sum {i=1}^N [(R{i} - r{i})/(t{i} ű
b{i})] or equivalently: sum {i=1}^N [(R{i} - r{i})/r{i}] with R{i} b{i})] or equivalently: sum {i=1}^N [(R{i} - r{i})/r{i}] with R{i}
>= 1 and r{i} >= 1 (in terms of per component bandwidth unit). In >= 1 and r{i} >= 1 (in terms of per component bandwidth unit). In
this formula, N is the total number of links traversed by a given this formula, N is the total number of links traversed by a given
LSP, t[i] the Maximum Bandwidth per TE link i and b[i] the sum per LSP, t[i] the Maximum Bandwidth per TE link i and b[i] the sum per
TE link i of the bandwidth committed for working LSPs and other TE link i of the bandwidth committed for working LSPs and other
recovery LSPs (thus except ˘shared bandwidth÷ LSPs). The quantity recovery LSPs (thus except "shared bandwidth" LSPs). The quantity
[(R{i} - r{i})/r{i}] is defined as the Shared (Recovery) Bandwidth [(R{i} - r{i})/r{i}] is defined as the Shared (Recovery) Bandwidth
Ratio per TE link i. In addition, TE links for which R[i] reaches Ratio per TE link i. In addition, TE links for which R[i] reaches
max_R[i] or for which r[i] = 0 are pruned during shared recovery max_R[i] or for which r[i] = 0 are pruned during shared recovery
D.Papadimitriou et al. - Internet Draft - Expires November 2003 31
path computation as well as TE links for which max_R[i] = r[i] which path computation as well as TE links for which max_R[i] = r[i] which
can simply not be shared. can simply not be shared.
More generally, one can draw the following mapping between the More generally, one can draw the following mapping between the
available bandwidth at the transport and control plane level: available bandwidth at the transport and control plane level:
- ---------- Max Reservable Bandwidth - ---------- Max Reservable Bandwidth
| ----- ^ | ----- ^
|R ----- | |R ----- |
| ----- | | ----- |
skipping to change at line 1671 skipping to change at line 1701
-------- TE link Capacity - ------ | - Maximum Bandwidth -------- TE link Capacity - ------ | - Maximum Bandwidth
----- |r ----- v ----- |r ----- v
----- <------ b ------> - ---------- Unreserved Bandwidth ----- <------ b ------> - ---------- Unreserved Bandwidth
----- ----- ----- -----
----- ----- ----- -----
----- ----- ----- -----
----- ----- ----- -----
----- ----- <--- Min LSP Bandwidth ----- ----- <--- Min LSP Bandwidth
-------- 0 ---------- 0 -------- 0 ---------- 0
D.Papadimitriou et al. - Internet Draft ű June 2003 31
Note that the above approach does not require the flooding of any Note that the above approach does not require the flooding of any
per LSP information or a detailed distribution of the bandwidth per LSP information or a detailed distribution of the bandwidth
allocation per component link (or individual ports). Moreover, it allocation per component link (or individual ports). Such approach
has been demonstrated that this Partial Information Routing approach is referred to as a Partial Information Routing approach where per-
can also be extended to resource shareability with respect to the priority bandwidth TE Link advertisements allow for the same
number of times each SRLG is protected by a recovery resource, in capability as if a dedicated unreserved recovery bandwidth sub-TLV
particular an LSP (see also Section 8.4.2). This method also was defined (as suggested in [KODIALAM]). The latter shows that the
referred to as stochastic approach is described in [BOUILLET]. By difference obtained with a Full Information Routing approach (where
flooding this summarized information using a link-state protocol, the set of working and recovery LSPs using a given link is known at
recovery path computation and selection for SRLG diverse recovery each node) is fairly close.
LSPs can be optimized with respect to resource sharing giving a
performance difference of less than 5% compared to a Full Moreover, it has also been demonstrated that the Partial Information
Information Flooding approach. The latter can be found in [GLI] for Routing approach can be extended to resource shareability with
instance. Note that strictly speaking both methods rely on respect to the number of times each SRLG is protected by a recovery
deterministic knowledge of the network topology and resource (usage) resource, in particular an LSP (see also Section 8.4.2). This
status. extended method is described in [BOUILLET]. By flooding this
aggregated information using a link-state routing protocol, recovery
path computation and selection for SRLG diverse recovery LSPs can be
optimized with respect to resource sharing giving a performance
difference of less than 5% (and so negligible) compared to a Full
Information Flooding approach. The latter is detailed in [GLI], for
instance. Note also that all these methods rely on deterministic
knowledge (at different degrees) of the network topology and
resource usage status.
For GMPLS-based recovery purposes, the Partial Information Routing For GMPLS-based recovery purposes, the Partial Information Routing
approach can be further enhanced by extending GMPLS signalling approach can be further enhanced by extending GMPLS signalling
capabilities. This, by allowing the working LSP related information capabilities. This, by allowing the working LSP related information
and in particular, its explicit route to be exchanged over the and in particular, its explicit route to be exchanged over the
recovery LSP in order to enable more efficient admission control at recovery LSP in order to enable more efficient admission control at
shared (link) resource upstream nodes. ingress nodes of shared resources, in particular links.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 32
8.4.2 Recovery Resource Sharing and SRLG Recovery 8.4.2 Recovery Resource Sharing and SRLG Recovery
As stated in the previous section, resource shareability can also be As stated in the previous section, resource shareability can also be
maximized with respect to the number of times each SRLG is protected maximized with respect to the number of times each SRLG is protected
by a recovery resource. by a recovery resource.
Methods can be considered for avoiding contention for the shared Methods can be considered for avoiding contention for the shared
recovery resources during a single SRLG failure (see Section 5). recovery resources during a single SRLG failure (see Section 5).
These allow the sharing of common reserved recovery resource between These allow the sharing of common reserved recovery resource between
skipping to change at line 1723 skipping to change at line 1762
sharing on a TE link such as the current number of recovery LSPs sharing on a TE link such as the current number of recovery LSPs
sharing the recovery resources (pre-)allocated on the TE link (see sharing the recovery resources (pre-)allocated on the TE link (see
also Section 8.4.1) and the current number of SRLGs recoverable by also Section 8.4.1) and the current number of SRLGs recoverable by
this amount of shared recovery resource on this TE link, may be this amount of shared recovery resource on this TE link, may be
considered. The latter is equivalent to the total number of SRLGs considered. The latter is equivalent to the total number of SRLGs
that the (recovery) LSPs sharing the recovery resources shall that the (recovery) LSPs sharing the recovery resources shall
recover. Then, if SRLG recoverability is considered, the explicit recover. Then, if SRLG recoverability is considered, the explicit
list of SRLGs recoverable by the recovery resources shared on the TE list of SRLGs recoverable by the recovery resources shared on the TE
link together with their respective sharable recovery bandwidth (see link together with their respective sharable recovery bandwidth (see
also Section 8.4.1) may be considered. The latter information is also Section 8.4.1) may be considered. The latter information is
equivalent to the maximum sharable recovery bandwidth per SRLG or equivalent to the maximum sharable recovery bandwidth per SRLG (or
per group of SRLG) which implies to consider a decreasing amount of
D.Papadimitriou et al. - Internet Draft ű June 2003 32
per group of SRLG implying to consider a decreasing amount of
sharable bandwidth and SRLG list over time. sharable bandwidth and SRLG list over time.
Compared to the case of simple recovery resource sharing regardless Compared to the case of recovery resource sharing only regardless of
of SRLG recoverability (as described in Section 8.4.1), the SRLG recoverability (as described in Section 8.4.1), the additional
additional TE link information considered here would potentially TE link information considered here would potentially allow for
allow for better path computation and selection (at distinct ingress better path computation and selection (at distinct ingress node)
node) during SRLG-disjoint LSP provisioning in a shared meshed during SRLG-disjoint LSP provisioning in a shared meshed recovery
recovery scheme. However, due to the lack of results of evidence for scheme. However, due to the lack of results of evidence for better
better efficiency and due to the complexity that such extensions efficiency (see also Section 8.4.1) and due to the complexity that
would in turn generate, these are not considered in the scope of such extensions would in turn generate, these extensions are not
the present analysis. For instance, a per (group of) SRLG maximum further considered in the scope of the present analysis. For
sharable recovery bandwidth is restricted by the length that the instance, a per (group of) SRLG maximum shareable recovery bandwidth
corresponding (sub-)TLV may take and thus the number of SRLGs that is restricted by the length that the corresponding (sub-)TLV may
it can include. Therefore, they SHOULD not be translated into GMPLS take and thus the number of SRLGs that it can include. Therefore,
routing or signalling protocol extensions for recovery purposes. the corresponding parameters SHOULD not be translated into GMPLS
routing (or even signalling) protocol extensions for recovery
purposes.
The next section will demonstrate that such extensions complements However, the next section will demonstrate that the exchange of the
the exchange of the explicit route of working LSP over the recovery path (including link and node identifiers) of the working LSP over
LSP path in order to achieve shared recovery resources contention the recovery LSP path helps in achieving shared recovery resources
avoidance. admission control.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 33
8.4.3 Recovery Resource Sharing, SRLG Disjointness and Admission 8.4.3 Recovery Resource Sharing, SRLG Disjointness and Admission
Control Control
Admission control is a strict requirement to be fulfilled by nodes Admission control is a strict requirement to be fulfilled by nodes
giving access to shared links. This can be illustrated using the giving access to shared links. This can be illustrated using the
following network topology: following network topology:
A ------ C ====== D A ------ C ====== D
| | | | | |
skipping to change at line 1774 skipping to change at line 1815
simultaneously a working LSP to D through C and a recovery LSP simultaneously a working LSP to D through C and a recovery LSP
(through E and F) to the same destination. Then, A decides to create (through E and F) to the same destination. Then, A decides to create
a recovery LSP to D, but since C to D span carries both working LSPs a recovery LSP to D, but since C to D span carries both working LSPs
node E should either assign a dedicated resource for this recovery node E should either assign a dedicated resource for this recovery
LSP or if it has already reached its maximum shared recovery LSP or if it has already reached its maximum shared recovery
bandwidth level reject this request. Otherwise, in the latter case a bandwidth level reject this request. Otherwise, in the latter case a
C-D span failure would imply that one of the working LSP would not C-D span failure would imply that one of the working LSP would not
be recoverable. be recoverable.
Consequently, node E must have the required information (implying Consequently, node E must have the required information (implying
for instance that the explicit route followed by the primary LSPs to for instance, that the explicit route followed by the working LSPs
be carried with the corresponding recovery LSP request) in order to to be carried with the corresponding recovery LSP request) in order
perform an admission control for the recovery LSP requests. to perform an admission control for the recovery LSP requests.
D.Papadimitriou et al. - Internet Draft ű June 2003 33
Moreover, node E may securely (if its maximum shared recovery Moreover, node E may securely (if its maximum shared recovery
bandwidth ratio has not been reached yet for this link) accept the bandwidth ratio has not been reached yet for this link) accept the
recovery LSP request and logically assign the same resource to these recovery LSP request and logically assign the same resource to these
LSPs. This if and only if it can guarantee that A-C-D and B-C-D are LSPs. This if and only if it can guarantee that A-C-D and B-C-D are
SRLG disjoint over the C-D span (one considers here in the scope of SRLG disjoint over the C-D span (one considers here in the scope of
this example, node failure probability as negligible). To achieve this example, node failure probability as negligible). To achieve
this, the explicit route of the primary LSP (and transported over this, the explicit route of the working LSP (and transported over
the recovery path) is examined at each shared link ingress node. The the recovery path) is examined at each shared link ingress node. The
latter uses the interface identifier as index to retrieve in the TE latter uses the interface identifier as index to retrieve in the TE
link State DataBase (TE LSDB) the SRLG id list associated to the link State DataBase (TE LSDB) the SRLG id list associated to the
links of the working LSPs. If these LSPs have one or more SRLG id in links of the working LSPs. If these LSPs have one or more SRLG id in
common (in this example, one or more SRLG id in common over C-D), common (in this example, one or more SRLG id in common over C-D),
then node E should not assign the same resource to the recovery then node E should not assign the same resource to the recovery
LSPs. Otherwise one of these working LSPs would not be recoverable LSPs. Otherwise one of these working LSPs would not be recoverable
in case of C-D span failure. in case of C-D span failure.
There are some issues related to this method, the major one being There are some issues related to this method, the major one being
the number of SRLG Ids that a single link can cover (more than 100, the number of SRLG Ids that a single link can cover (more than 100,
in complex environments). Moreover, when using link bundles, this in complex environments). Moreover, when using link bundles, this
approach may generate the rejection of some recovery LSP requests. approach may generate the rejection of some recovery LSP requests.
This because the SRLG sub-TLV corresponding to a link bundle This because the SRLG sub-TLV corresponding to a link bundle
includes the union of the SRLG id list of all the component links includes the union of the SRLG id list of all the component links
belonging to this bundle (see [GMPLS-RTG] and [MPLS-BUNDLE]). belonging to this bundle (see [GMPLS-RTG] and [MPLS-BUNDLE]).
D.Papadimitriou et al. - Internet Draft - Expires November 2003 34
In order to overcome this specific issue, an additional mechanism In order to overcome this specific issue, an additional mechanism
may consist of querying the nodes where such an information would be may consist of querying the nodes where such an information would be
available (in this case, node E would query C). The major drawback available (in this case, node E would query C). The main drawback of
of this method is, in addition to the dedicated mechanism it this method is that, in addition to the dedicated mechanism(s) it
requires, that it may become very complex when several common nodes requires, it may become complex when several common nodes are
are traversed by the working LSPĂs. Therefore, when using link traversed by the working LSPs. Therefore, when using link bundles,
bundles, a potential way of solving this issue tightly related to solving this issue (tightly related to the sequence of the recovery
the sequence of the recovery operations (at least in a first step, operations and since per component flooding of SRLG identifiers
since per component flooding of SRLG identifiers would impact the would impact the link state routing protocol scalability), may rely
link state routing protocol scalability), is to rely on the usage of on the usage of an on-line accessible network management system.
an on-line accessible network management system.
9. Summary 9. Summary and Conclusions
One can summarize by the following table the selection of a recovery The following table summarizes the different recovery types and
scheme/strategy, using the recovery types proposed in [CCAMP-TERM] schemes analyzed throughout this document.
and their detailed analysis proposed in this memo.
-------------------------------------------------------------------- --------------------------------------------------------------------
| Path Search (computation and selection) | Path Search (computation and selection)
-------------------------------------------------------------------- --------------------------------------------------------------------
| Pre-planned (a) | Dynamic (b) | Pre-planned (a) | Dynamic (b)
-------------------------------------------------------------------- --------------------------------------------------------------------
| | faster recovery | Does not apply | | faster recovery | Does not apply
| | less flexible | | | less flexible |
| 1 | less robust | | 1 | less robust |
| | most resource consuming | | | most resource consuming |
Path | | | Path | | |
Setup ------------------------------------------------------------ Setup ------------------------------------------------------------
D.Papadimitriou et al. - Internet Draft ű June 2003 34
| | relatively fast recovery | Does not apply | | relatively fast recovery | Does not apply
| | relatively flexible | | | relatively flexible |
| 2 | relatively robust | | 2 | relatively robust |
| | resource consumption | | | resource consumption |
| | depends on sharing degree | | | depends on sharing degree |
------------------------------------------------------------ ------------------------------------------------------------
| | relatively fast recovery | less faster (computation) | | relatively fast recovery | less faster (computation)
| | more flexible | most flexible | | more flexible | most flexible
| 3 | relatively robust | most robust | 3 | relatively robust | most robust
| | less resource consuming | least resource consuming | | less resource consuming | least resource consuming
| | depends on sharing degree | | | depends on sharing degree |
-------------------------------------------------------------------- --------------------------------------------------------------------
1. Recovery LSP setup (before failure occurrence) with resource 1a. Recovery LSP setup (before failure occurrence) with resource
reservation (i.e. signalling) and selection is referred to as LSP reservation (i.e. signalling) and selection is referred to as
protection. LSP protection.
2. Recovery LSP setup (before failure occurrence) with resource 2a. Recovery LSP setup (before failure occurrence) with resource
reservation (i.e. signalling) with resource pre-selection is reservation (i.e. signalling) and with resource pre-selection is
referred to as pre-planned LSP re-routing with resource pre- referred to as pre-planned LSP re-routing with resource pre-
selection. This implies only recovery LSP activation after selection. This implies only recovery LSP activation after
failure occurrence. failure occurrence.
3a. Recovery LSP setup (before failure occurrence) with resource 3a. Recovery LSP setup (before failure occurrence) with resource
reservation (i.e. signalling) without resource selection is reservation (i.e. signalling) and without resource selection is
D.Papadimitriou et al. - Internet Draft - Expires November 2003 35
referred to as pre-planned LSP re-routing without resource pre- referred to as pre-planned LSP re-routing without resource pre-
selection. This implies recovery LSP activation and resource selection. This implies recovery LSP activation and resource
(i.e. label) selection after failure occurrence. (i.e. label) selection after failure occurrence.
3b. Recovery LSP (full) setup after failure occurrence is referred 3b. Recovery LSP setup after failure occurrence is referred to as
to as dynamic LSP re-routing. to as LSP re-routing, which is full when recovery LSP path
computation occurs after failure occurrence.
Thus, the term pre-planned refers to recovery resource pre- Thus, the term pre-planned refers here to recovery resource pre-
computation, signaling (reservation) and a priori selection computation, signaling (reservation) and a priori selection
(optional), but not cross-connection. (optional), but not cross-connection. Also, the shared-mesh recovery
scheme can be view as a particular case of 2a) and 3a) using the
additional constraint described in section 8.4.3.
The implementation of these recovery mechanisms and their
corresponding phases requires only extensions to GMPLS signalling
protocols (i.e. [RFC3471] and [RFC3473]). The present analysis
demonstrates (in Section 8) that no GMPLS routing extensions are
expected in order for GMPLS to provide any of these recovery types
and schemes. These GMPLS signalling extensions should mainly focus
in delivering 1) recovery LSP pre-provisioning (only for the cases
1a, 2a and 3a) 2) failure notification 3) recovery switching actions
and 4) reversion mechanisms.
10. Security Considerations 10. Security Considerations
This document does not introduce or imply any specific security This document does not introduce or imply any specific security
consideration. consideration.
11. References 11. Acknowledgments
11.1 Normative References The authors would like to thank Fabrice Poppe (Alcatel) and Bart
Rousseau (Alcatel) for their revision effort, Richard Rabbat
(Fujitsu), David Griffith (NIST) and Lyndon Ong (Ciena) for their
useful comments.
[GMPLS-ARCH] E.Mannie (Editor), ˘Generalized MPLS Architecture÷, 12. Intellectual Property Considerations
This section is taken from Section 10.4 of [RFC2026].
The IETF takes no position regarding the validity or scope of any
intellectual property or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; neither does it represent that it
has made any effort to identify any such rights. Information on the
IETF's procedures with respect to rights in standards-track and
standards-related documentation can be found in BCP-11. Copies of
claims of rights made available for publication and any assurances
of licenses to be made available, or the result of an attempt made
to obtain a general license or permission for the use of such
proprietary rights by implementors or users of this specification
can be obtained from the IETF Secretariat.
D.Papadimitriou et al. - Internet Draft - Expires November 2003 36
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights which may cover technology that may be required to practice
this standard. Please address the information to the IETF Executive
Director.
13. References
13.1 Normative References
[CCAMP-TERM] E.Mannie and D.Papadimitriou (Editors), "Recovery
(Protection and Restoration) Terminology for GMPLS,"
Internet Draft, Work in progress, draft-ietf-ccamp- Internet Draft, Work in progress, draft-ietf-ccamp-
gmpls-architecture-03.txt, August 2002. gmpls-recovery-terminology-02.txt, May 2003.
[GMPLS-RTG] K.Kompella (Editor), ˘Routing Extensions in Support of [GMPLS-ARCH] E.Mannie (Editor) et al., "Generalized MPLS
Generalized MPLS,÷ Internet Draft, Work in Progress, Architecture," Work in progress, draft-ietf-ccamp-
draft-ietf-ccamp-gmpls-routing-05.txt, August 2002. gmpls-architecture-07.txt, May 2003.
D.Papadimitriou et al. - Internet Draft ű June 2003 35 [GMPLS-RTG] K.Kompella (Editor) et al., "Routing Extensions in
[GMPLS-SIG] L.Berger (Editor), ˘Generalized MPLS ű Signaling Support of Generalized MPLS," Work in Progress, draft-
Functional Description÷, Internet Draft, Work in ietf-ccamp-gmpls-routing-05.txt, August 2002.
progress, draft-ietf-mpls-generalized-signaling-09.txt,
October 2002.
[LMP] J.Lang (Editor), ˘Link Management Protocol (LMP) v1.0÷ [LMP] J.P.Lang (Editor) et al., "Link Management Protocol
Internet Draft, Work in progress, draft-ietf-ccamp-lmp- (LMP) v1.0," Internet Draft, Work in progress, draft-
07, October 2002. ietf-ccamp-lmp-09.txt, May 2003.
[LMP-WDM] A.Fredette and J.Lang (Editors), ˘Link Management [LMP-WDM] A.Fredette and J.P.Lang (Editors), "Link Management
Protocol (LMP) for DWDM Optical Line Systems,÷ Internet Protocol (LMP) for DWDM Optical Line Systems," Work in
Draft, Work in progress, draft-ietf-ccamp-lmp-wdm- progress, draft-ietf-ccamp-lmp-wdm-02.txt, March 2003.
01.txt, September 2002.
11.2 Informative References [RFC-2026] S.Bradner, "The Internet Standards Process -- Revision
3," BCP 9, IETF RFC 2026, October 1996.
[RFC-2026] Bradner, S., ˘The Internet Standards Process -- [RFC-2119] S.Bradner, "Key words for use in RFCs to Indicate
Revision 3÷, BCP 9, RFC 2026, October 1996. Requirement Levels," BCP 14, IETF RFC 2119, March 1997.
[RFC-2119] Bradner, S., ˘Key words for use in RFCs to Indicate [RFC-3471] L.Berger (Editor) et al., "Generalized MPLS - Signaling
Requirement Levels÷, BCP 14, RFC 2119, March 1997. Functional Description," IETF RFC 3471, January 2003.
[BOUILLET] E.Bouillet et al., ˘Stochastic Approaches to Compute [RFC-3473] L.Berger (Editor) et al., "Generalized MPLS Signaling -
RSVP-TE Extensions," IETF RFC 3473, January 2003.
13.2 Informative References
[BOUILLET] E.Bouillet et al., "Stochastic Approaches to Compute
Shared Meshed Restored Lightpaths in Optical Network Shared Meshed Restored Lightpaths in Optical Network
Architectures÷, INFOCOM 2002, New York City, June 2002. Architectures," IEEE Infocom 2002, New York City, June
2002.
[CCAMP-LI] G.Li et al. ˘RSVP-TE Extensions For Shared-Mesh [CCAMP-LI] G.Li et al. "RSVP-TE Extensions For Shared-Mesh
Restoration in Transport Networks÷, Internet Draft, Restoration in Transport Networks," Internet Draft,
D.Papadimitriou et al. - Internet Draft - Expires November 2003 37
Work in progress, draft-li-shared-mesh-restoration- Work in progress, draft-li-shared-mesh-restoration-
01.txt, November 2001. 01.txt, November 2001.
[CCAMP-LIU] H.Liu et al. ˘OSPF-TE Extensions in Support of Shared [CCAMP-LIU] H.Liu et al. "OSPF-TE Extensions in Support of Shared
Mesh Restoration÷, Internet Draft, Work in progress, Mesh Restoration," Internet Draft, Work in progress,
draft-liu-gmpls-ospf-restoration-00.txt, October 2002. draft-liu-gmpls-ospf-restoration-00.txt, October 2002.
[CCAMP-SRLG] D.Papadimitriou et al., ˘Shared Risk Link Groups [CCAMP-SRLG] D.Papadimitriou et al., "Shared Risk Link Groups
Encoding and Processing,÷ Internet Draft, Work in Encoding and Processing," Internet Draft, Work in
progress, draft-papadimitriou-ccamp-srlg-processing- progress, draft-papadimitriou-ccamp-srlg-processing-
01.txt, November 2002. 01.txt, November 2002.
[CCAMP-TERM] E.Mannie and D.Papadimitriou (Editors), ˘Recovery [DEMEESTER] P.Demeester et al., "Resilience in Multilayer
(Protection and Restoration) Terminology for GMPLS,÷ Networks," IEEE Communications Magazine, Vol. 37, No.
Internet Draft, Work in progress, draft-ietf-ccamp- 8, pp. 70-76, August 1998.
gmpls-recovery-terminology-00.txt, June 2002.
[DEMEESTER] P.Demeester et al., ˘Resilience in Multilayer
Networks÷, IEEE Communications Magazine, Vol. 37, No.
8, August 1998, pp. 70-76.
[G.707] ITU-T, ˘Network Node Interface for the Synchronous [G.707] ITU-T, "Network Node Interface for the Synchronous
Digital Hierarchy (SDH)÷, Recommendation G.707, October Digital Hierarchy (SDH)," Recommendation G.707, October
2000. 2000.
D.Papadimitriou et al. - Internet Draft ű June 2003 36 [G.709] ITU-T, "Network Node Interface for the Optical
[G.709] ITU-T, ˘Network Node Interface for the Optical Transport Network (OTN)," Recommendation G.709,
Transport Network (OTN)÷, Recommendation G.709,
February 2001 (and Amendment nŚ1, October 2001). February 2001 (and Amendment nŚ1, October 2001).
[G.783] ITU-T, ˘Characteristics of Synchronous Digital [G.783] ITU-T, "Characteristics of Synchronous Digital
Hierarchy (SDH) Equipment Functional Blocks÷, Hierarchy (SDH) Equipment Functional Blocks,"
Recommendation G.783, October 2000. Recommendation G.783, October 2000.
[G.798] ITU-T, ˘Characteristics of Optical Transport Network [G.798] ITU-T, "Characteristics of Optical Transport Network
(OTN) Equipment Functional Blocks÷, Recommendation (OTN) Equipment Functional Blocks," Recommendation
G.798, January 2002. G.798, January 2002.
[G.806] ITU-T, ˘Characteristics of Transport Equipment ű [G.806] ITU-T, "Characteristics of Transport Equipment ű
Description Methodology and Generic Functionality÷, Description Methodology and Generic Functionality",
Recommendation G.806, October 2000. Recommendation G.806, October 2000.
[G.826] ITU-T, ˘Performance Monitoring÷, Recommendation G.826, [G.826] ITU-T, "Performance Monitoring," Recommendation G.826,
February 1999. February 1999.
[G.841] ITU-T, ˘Types and Characteristics of SDH Network [G.808.1] ITU-T, "Generic Protection Switching ű Linear trail and
Protection Architectures÷, Recommendation G.841, Subnetwork Protection," Draft Recommendation (work in
progress), Version 0.5, January 2003.
[G.841] ITU-T, "Types and Characteristics of SDH Network
Protection Architectures," Recommendation G.841,
October 1998. October 1998.
[G.842] ITU-T, ˘Interworking of SDH network protection [G.842] ITU-T, "Interworking of SDH network protection
architectures÷, Recommendation G.842, October 1998. architectures," Recommendation G.842, October 1998.
[G.GPS] ITU-T Draft Recommendation G.GPS, Version 2, ˘Generic [GLI] G.Li et al., "Efficient Distributed Path Selection for
Protection Switching÷, Work in progress, May 2002. Shared Restoration Connections," IEEE Infocom 2002, New
York City, June 2002.
[GLI] Guangzhi Li et al., ˘Efficient Distributed Path D.Papadimitriou et al. - Internet Draft - Expires November 2003 38
Selection for Shared Restoration Connections÷, IEEE [KODIALAM] M.Kodialam and T.V.Lakshman, "Restorable Dynamic
Infocom, New York, June 2002. Quality of Service Routing," IEEE Communications
Magazine, pp. 72-81, June 2002.
[MANCHESTER] J.Manchester, P.Bonenfant and C.Newton, ˘The Evolution [MANCHESTER] J.Manchester, P.Bonenfant and C.Newton, "The Evolution
of Transport Network Survivability,÷ IEEE of Transport Network Survivability," IEEE
Communications Magazine, August 1999. Communications Magazine, August 1999.
[MPLS-REC] V.Sharma and F.Hellstrand (Editors) et al., ˘A [MPLS-OSU] S.Seetharaman et al., "IP over Optical Networks: A
Framework for MPLS Recovery÷, Internet Draft, Work in Summary of Issues," Internet Draft, Work in Progress,
Progress, draft-ietf-mpls-recovery-frmwrk-06.txt, July draft-osu-ipo-mpls-issues-02.txt, April 2001.
[RFC-3386] W.Lai, D.McDysan, J.Boyle, et al., "Network Hierarchy
and Multi-layer Survivability," IETF RFC 3386, November
2002. 2002.
[MPLS-OSU] S.Seetharaman et al, ˘IP over Optical Networks: A [RFC-3469] V. Sharma and F. Hellstrand (Editors), "Framework for
Summary of Issues÷, Internet Draft, Work in Progress, Multi-Protocol Label Switching (MPLS)- based Recovery,"
draft-osu-ipo-mpls-issues-02.txt, April 2001. IETF RFC 3469, February 2003.
[T1.105] ANSI, "Synchronous Optical Network (SONET): Basic [T1.105] ANSI, "Synchronous Optical Network (SONET): Basic
Description Including Multiplex Structure, Rates, and Description Including Multiplex Structure, Rates, and
Formats", ANSI T1.105, January 2001. Formats," ANSI T1.105, January 2001.
[TE-NS] K.Owens et al, ˘Network Survivability Considerations
for Traffic Engineered IP Networks÷, Internet Draft,
D.Papadimitriou et al. - Internet Draft ű June 2003 37 [TE-NS] K.Owens et al., "Network Survivability Considerations
for Traffic Engineered IP Networks," Internet Draft,
Work in Progress, draft-owens-te-network-survivability- Work in Progress, draft-owens-te-network-survivability-
01.txt, July 2001. 01.txt, July 2001.
[TE-RH] W.Lai, D.McDysan, J.Boyle, et al, ˘Network Hierarchy [WANG] J.Wang, L.Sahasrabuddhe, and B.Mukherjee, "Path vs.
and Multi-layer Survivability÷, Internet Draft, Work in Subpath vs. Link Restoration for Fault Management in
Progress, draft-ietf-tewg-restore-hierarchy-01.txt, IP-over-WDM Networks: Performance Comparisons Using
June 2002. GMPLS Control Signaling," IEEE Communications Magazine,
pp. 80-87, November 2002.
12. Acknowledgments
The authors would like to thank Fabrice Poppe (Alcatel) and Bart
Rousseau (Alcatel) for their revision effort, Richard Rabbat
(Fujitsu), David Griffith (NIST) and Lyndon Ong (Ciena) for their
useful comments.
13. Author's Addresses 14. Author's Addresses
Eric Mannie (Consult) Eric Mannie (Consult)
E-mail: eric_mannie@hotmail.com E-mail: eric_mannie@hotmail.com
Dimitri Papadimitriou (Alcatel) Dimitri Papadimitriou (Alcatel)
Francis Wellesplein, 1 Francis Wellesplein, 1
B-2018 Antwerpen, Belgium B-2018 Antwerpen, Belgium
Phone : +32 3 240-8491 Phone : +32 3 240-8491
E-mail: dimitri.papadimitriou@alcatel.be E-mail: dimitri.papadimitriou@alcatel.be
D.Papadimitriou et al. - Internet Draft ű June 2003 38 D.Papadimitriou et al. - Internet Draft - Expires November 2003 39
Full Copyright Statement Full Copyright Statement
"Copyright (C) The Internet Society (date). All Rights Reserved. "Copyright (C) The Internet Society (date). All Rights Reserved.
This document and translations of it may be copied and furnished to This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it others, and derivative works that comment on or otherwise explain it
or assist in its implementation may be prepared, copied, published or assist in its implementation may be prepared, copied, published
and distributed, in whole or in part, without restriction of any and distributed, in whole or in part, without restriction of any
kind, provided that the above copyright notice and this paragraph kind, provided that the above copyright notice and this paragraph
are included on all such copies and derivative works. However, this are included on all such copies and derivative works. However, this
skipping to change at line 2051 skipping to change at line 2137
The limited permissions granted above are perpetual and will not be The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns. revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an This document and the information contained herein is provided on an
"AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE." MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE."
D.Papadimitriou et al. - Internet Draft ű June 2003 39 D.Papadimitriou et al. - Internet Draft - Expires November 2003 40
 End of changes. 

This html diff was produced by rfcdiff 1.23, available from http://www.levkowetz.com/ietf/tools/rfcdiff/