[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: draft-bernstein-ccamp-gmpls-vcat-lcas-01 comments - 3rd try
- To: ccamp <ccamp@ops.ietf.org>
- Subject: Re: draft-bernstein-ccamp-gmpls-vcat-lcas-01 comments - 3rd try
- From: Huub van Helvoort <hhelvoort@chello.nl>
- Date: Sun, 23 Oct 2005 11:30:45 +0200
- User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.2) Gecko/20040804 Netscape/7.2 (ax)
Thanks Adrian,
Huub, I don't see them in the spam bucket, but given the rate of arrival
of new spam it is possible they went out the over flow before I got there
:-(
Please post again.
OK, the resend with attachment did not work, this time in-line
without diff-marks.
My comments are based on ITU-T recommendation G.806 clause 10
and appendix VII.
This may be a good reference to add at the end of the draft.
If you need more information, please feel free to ask me or
read all about VCAT, LCAS (and GFP) in my book:
"Next generation SDH/SONET: Evolution or Revolution?"
John Wiley & Sons ISBN: 0-470-09120-7
Cheers, Huub.
================================================================
http://members.chello.nl/hhelvoort
================================================================
Always remember that you are unique...just like everyone else...
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
CCAMP working Group G. Bernstein
Internet-Draft Grotto Networking
Expires: April 14, 2006 D. Caviglia
Marconi
R. Rabbat
Fujitsu
October 11, 2005
Operating Virtual concatenation (VCAT) and the Link Capacity Adjustment
Scheme (LCAS) with Generalized Multi-Protocol Label Switching (GMPLS)
draft-bernstein-ccamp-gmpls-vcat-lcas-01
Status of this Memo
By submitting this Internet-Draft, each author represents that any
applicable patent or other IPR claims of which he or she is aware
have been or will be disclosed, and any of which he or she becomes
aware will be disclosed, in accordance with Section 6 of BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This Internet-Draft will expire on April 14, 2006.
Copyright Notice
Copyright (C) The Internet Society (2005).
Abstract
This document describes the use of the Generalized Multi-Protocol
Label Switching (GMPLS) control plane in conjunction with the Virtual
Concatenation (VCAT) layer 1 inverse multiplexing mechanism and its
companion Link Capacity Adjustment Scheme (LCAS) which can be used
Bernstein, et al. Expires April 14, 2006 [Page 1]
Internet-Draft GMPLS, VCAT and LCAS October 2005
for hitless dynamic resizing of the inverse multiplex group. These
techniques apply to the Optical Transport Network (OTN), Synchronous
Optical Network (SONET), Synchronous Digital Hierarchy (SDH) and
Plesiochronous Digital Hierarchy (PDH) signals.
Table of Contents
1. Overview of VCAT and LCAS . . . . . . . . . . . . . . . . . . 3
1.1. VCAT signals and components . . . . . . . . . . . . . . . 3
1.2. VCAT Capabilities and Limitations . . . . . . . . . . . . 3
1.3. The LCAS Protocol . . . . . . . . . . . . . . . . . . . . 4
2. Problem Statement and Current Support . . . . . . . . . . . . 5
2.1. Discovery of Enabled End Systems . . . . . . . . . . . . . 5
2.2. Client to End Point Mappings . . . . . . . . . . . . . . . 5
2.3. VCAT configuration without LCAS . . . . . . . . . . . . . 6
2.4. VCAT configuration with LCAS . . . . . . . . . . . . . . . 7
2.5. Component Signal Configuration Scenarios . . . . . . . . . 8
3. Possible Extensions to GMPLS to support additional
VCAT/LCAS scenarios . . . . . . . . . . . . . . . . . . . . . 10
3.1. Mechanisms for Discovery of VCAT/LCAS . . . . . . . . . . 10
3.2. Mechanism to Support Multiple Client to End Point
Mappings . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.3. Support for Component Signal Configuration Scenarios . . . 10
4. References . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Appendix A. Acknowledgements . . . . . . . . . . . . . . . . . . 12
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 13
Intellectual Property and Copyright Statements . . . . . . . . . . 14
Bernstein, et al. Expires April 14, 2006 [Page 2]
Internet-Draft GMPLS, VCAT and LCAS October 2005
1. Overview of VCAT and LCAS
Virtual Concatenation (VCAT) is a standardized layer 1 inverse
multiplexing technique that can be applied to OTN [6], SONET [3], SDH
[2], and PDH [5] component signals. By inverse multiplexing we mean
a method that combines multiple links at a particular layer into an
aggregate link to achieve a commensurate bandwidth capacity on that
aggregate link. More formally, VCAT essentially combines the payload
bandwidth of multiple path layer network signals (or trails) to support
a single client (e.g. Ethernet) layer link.
For a more detailed introduction see [1].
1.1. VCAT signals and components
In the following we will use SDH terminology rather than both SONET and
SDH terminology. In SDH Virtual Concatenation (VCAT) can be applied to
the following component time division multiplex (TDM) signals
referred to
as Virtual Containers (VCs): VC-11, VC-12, VC-2, VC-3, and VC-4.
Only like component signals can be aggregated into a VCAT group (VCG).
These VCGs are respectively known as: VC-11-Xv, VC-12-Xv, VC-2-Xv,
VC-3-Xv, and VC-4-Xv. In the previous designations X is an integer
indicating the number of members in the VCG. The value of X for high
order VCs (VC-3 and VC-4)can be from 1 to 256; and for low order VCs
(VC-11 VC-12 and VC-2) the value of X can be 1 to 64. See [2] for details.
VCAT can be applied to the following PDH signals as specified in
reference
[5]: DS1,E1, E3, DS3. Similar to the SONET/SDH case these component
signals can only be combined with like signals to produce
aggregates. For
some reason the VCGs of the PDH signals were not given unique
designations
in [5] so we shall adopt a similar notation to the SDH VCAT signals for
the permitted PDH VCAT signals that follow: DS1-Xv, E1-Xv, E3-Xv,
DS3-Xv.
Concatenation in the optical transport network (OTN) is realized by
means of virtual concatenation of Optical Channel Payload Unit (OPU)
signals. OPUk signals (k=1, 2, 3) can be concatenated into OPUk-Xv
aggregates with X= 1,..., 256. See reference [6] for details.
1.2. VCAT Capabilities and Limitations
VCAT performs inverse multiplexing by octet/byte de-interleaving of the
encapsulated client bit stream into the individual VCs of the VCG
and the
VCs are transported across the network independently from each
other. Due
to the possible different propagation times of the VCs, a differential
delay may occur between the individual VCs at the path termination. This
differential delay has to be compensated and the VCs have to be
realigned.
This is summarized for the different signal types in reference [1] with
details given in the respective standards documents [2] [3].
Bernstein, et al. Expires April 14, 2006 [Page 3]
Internet-Draft GMPLS, VCAT and LCAS October 2005
1.3. The LCAS Protocol
The Link Capacity Adjustment Scheme for VCAT signals is a protocol
for dynamically and hitlessly changing (i.e., increasing and
decreasing) the capacity of a VCG. LCAS also provides survivability
capabilities, automatically decreasing the capacity if a member of the
VCG experiences a failure in the network, and increasing the capacity
when the network fault is repaired. LCAS, itself, provides a mechanism
for interworking between LCAS and non-LCAS VCAT end points. VCAT does
not require LCAS for its operation.
LCAS functionality does not overlap or conflict with GMPLS' routing or
signaling functionality for the establishment of component links or
entire VCGs. LCAS instead is used to control whether a particular
component signal is actually put into service carrying traffic for
the VCG.
LCAS provides for graceful degradation of failed links by having the
sink
end(s) report back the receive status of all member components. In the
case of a reported member failure, the source end will stop using the
failed member and distribute the data over the remaining ‘good’ members,
thus transmission is restored to a reduced bandwidth capacity. The
source
end will send an LCAS message to the sink end that it is not
transmitting
data on that failed member. The worst case notification times are
summarized in [1].
Bernstein, et al. Expires April 14, 2006 [Page 4]
Internet-Draft GMPLS, VCAT and LCAS October 2005
2. Problem Statement and Current Support
In this section we list a number of VCAT/LCAS usage scenarios and
their current level of support. We will evaluate the applicability
of GMPLS to these scenarios and for those scenarios that GMPLS does
not currently support we describe possible GMPLS extensions in
Section 3. Note the term "component" signal in the text is used as a
simplified notion to the more formal concepts of VC-n, ODUk, and PDH
termination function as well as VC-n, ODUk and PDH path/trail.
2.1. Discovery of Enabled End Systems
Discovering VCAT: VCAT sources can only communicate with VCAT capable
sinks. Hence the VCAT capabilities of a PDH, SDH, or OTN path
termination points need to be known.
Currently no support for discovery of VCAT or LCAS apriori, i.e.,
via routing information. Support for "discovery" of VCAT
capability at connection establishment time via signaling, i.e.,
we can request VCAT connection and if the end system cannot
support it,it would refuse the connection. TBD -- is there a
specific error code concerning "VCAT not supported".
Discovering LCAS: LCAS offers additional functionality between VCAT
capable sources and sinks. Hence the LCAS capabilities of VCAT
enabled path termination points can be useful to know in advance
of component signal setup.
Currently there is no mechanism to ask for an LCAS enabled end
point nor is there a way to find out if the other end is LCAS
enabled until after the connection is established. This is a
problem if we specifically want hitless dynamic resizing or fast
graceful degradation for a VCG.
2.2. Client to End Point Mappings
Fixed Client to End point Mapping: Per client signal there is a
VC-n-Xv circuit in which the X VC-n termination points are
dedicated to this client signal. At any point in time, Xa (active
members) out of Xp (provisioned members) VC-n termination points may
be set up to carry client traffic.
For example when VCAT is deployed on a Router, the VCG connects
directly to one STM-N interface port (in absence of a HO or LO switch
fabric in the router). The transport network may then transport the
individual VCs via one or more routes. Diverse routing will incur a
differtial delay.
Bernstein, et al. Expires April 14, 2006 [Page 5]
Internet-Draft GMPLS, VCAT and LCAS October 2005
Variable Client to End point Mapping: For a set of M client signals
there are M*VC-n-Xv VCAT endpoints sharing a set of N (N>M) VC-n
termination points. It is possible that M*X > N (example: M=10, X=7,
N=64); i.e. there may be a kind of overbooking. Implication: must be
able to accommodate multiple different sized VCGs at an "interface".
For example an STM-64 interface can support many different VC-4-Xv
groups, providing the total number of VCs does not exceed the
interface
total capacity.
Implications: In both these cases we can have more than one VCG per
GMPLS
link. In general it is the responsibility of the signaling entity at
the source and destination ends to choose the appropriate VC-n
termination points for the VCG and in the case of "variable client
mapping" to perform needed internal configuration. However multiple
VCGs per GMPLS link or GMPLS addressed entity introduces an
unresolvable ambiguity when disjoint connections are set up or
dynamic resizing is applied since there is currently no "VCG
identifier" in GMPLS signaling.
2.3. VCAT configuration without LCAS (MI_LCASEnable=false)
Base Configuration: For VCAT to operate the sink end needs to be
informed of how many components are in the VCG. It has no
other way of knowing if it is currently receiving all components
intended to be in the group.
Fixed sized co-routed groups are supported with current GMPLS
signaling. Diversely routed groups are not currently supported.
Group Resizing: Additions or removals of components from a non-LCAS VCG
are not hitless, that is data loss will occur while the source and
sink are reprovisioned as to the number of members in the group.
Currently not supported within GMPLS. In particular, with each
addition or removal of a component the sink end point needs to be
told the expected number of components in the group.
Failure Conditions: Failure of a component must detected external to
VCAT system. The entire group is rendered inoperable until the
failed
component is taken out of service (or replaced) by
re-provisioning both
source and sink to that effect.
Currently not supported within GMPLS. The source and sink need
to be told
what component has failed and remove it from service.
Bernstein, et al. Expires April 14, 2006 [Page 6]
Internet-Draft GMPLS, VCAT and LCAS October 2005
2.4. VCAT configuration with LCAS (MI_LCASEnable=true)
Base Configuration: Sink end (and source end) are first configured with
the value of Xp (the provisioned number of members), out of Xm (the
technology specific maximum number of members) with Xm ³ Xp. LCAS
then determines automatically the number of active members Xa and
puts them into service for the group.
NOTE: The value of Xm is not only technology dependent, it can also be
implementation dependent. E.g. in a routerinterface with STM-64 line
rate for a VC-4-Xv based VCG Xm = 64 and not 256. Xm cannot be
provisioned, it can be retreived from the VCG (as well as Xa).
Currently both co-routed and diversely routed VCGs can be
supported if there is no VCG ambiguity.
Component Addition: When a new component signal has to be added to a
VCG the following procedure applies.
1. Establish a connection from Source to Sink for the member to
be added.
2. Provision the Source to use the ‘new’ member as one of the active
group (Xp -> Xp+1). This entails sending the ADD command by the new member.
3. Provision the Sink to recognize the ‘new’ member of the VCG (Xp ->
Xp+1).
The order of doing the above is not important. When the ADD command is
received by the sink it will return an ‘OK’ to the Source at which time
the Source will use all active members (old + new) to transport the
client traffic.
This procedure does not affect the already established LCAS members,
that is, client traffic is not sent on the new component
until the
LCAS procedure is complete;
This can be supported within GMPLS if, after GMPLS has successfully
established a potential new component, the source end
LCAS is (internally) told to add it to the group.
Component Removal: When a component is removed the following procedure
applies:
1. On receipt of a remove command (for a particular member) from the
management system, LCAS Source will stop using that member; the Sink
will recognize that that member no longer carrier client traffic and
will use only the remaining members to reconstitute the client signal.
Note: The current LCAS protocol allows for the ‘decrease’ command to be
initiated at either the LCAS source or the LCAS sink; if it is initiated
at the sink there will be a hit to the reconstructed data.
2. The connection, source provisioning and sink provisioning of that
member can then be removed from the VCG.
Bernstein, et al. Expires April 14, 2006 [Page 7]
Internet-Draft GMPLS, VCAT and LCAS October 2005
3. The component connection can be, if needed, removed from the
transport network.
This can be supported within GMPLS if, before GMPLS tears down a
component, LCAS is told (local to source end) to remove the
component from service in the group.
Component Failure: When a component fails:
1. The LCAS sink detects the failure (how this is done is outside the
scope of this ID) and informs the source of this failure via the member
status (MST)information. :
2. The source then takes the failed component out of service and
uses only the remaining ‘good’ members to transport the client traffic
(Xa -> Xa-1). Note: No re-sequencing of the VCG is done at this time.
When the failed component is repaired, LCAS can automatically add
the repaired component back to the group, or alternatively a new
component can be added to bring the group back to its original
size. Note that component failure is not hitless, but note the
fast notification times of [BernDiegoRabbat].
Currently supported since no action required of GMPLS.
2.5. Component Signal Configuration Scenarios
Here we use the term "group" to refer to the entire VCG and
the terminology "set" and "subset" to refer to the collection of
potential VCG member signals. Note that all assesments of
whether a scenario is currently supported assumes either (a) a single
VCG per GMPLS addressable entity, or (b) a mechanism is in
place to disambiguate multiple VCGs (see Section 2.2).
Fixed, Co-routed: A fixed bandwidth VCG, with all members
transported over
the same route. This is the case where the
intended bandwidth of the VCG does not change and all
member signals follow the same route. The intent here is the
capability to allocate the "right" amount of bandwidth.
Currently supported in GMPLS.
Fixed, Diverse-routes: A fixed bandwidth VCG, with the members
transported
over at least two different routes. The intent here is additional
resilience and graceful degradation in the case of failure.
Bernstein, et al. Expires April 14, 2006 [Page 8]
Internet-Draft GMPLS, VCAT and LCAS October 2005
If LCAS is present this scenario is supported under GMPLS. Not
supported without LCAS since we need two-way communications of
some type between source and sink to coordinate which members are
to be used in the group in a failure scenario.
Dynamic, Co-routed: A dynamic VCG (bandwidth can be increased
or decreased via the addition or removal of member signals),
with all the members transported over the same route. Intent here is
dynamic sizing of bandwidth.
If LCAS present this scenario is currently supported by GMPLS.
Implications: LCAS is needed for hitless resizing. Note before
LCAS can do its part of getting traffic over the modified VCG,
the two VCAT/LCAS endpoints need to be configured (Xp -> Xp+1 or Xp ->
Xp-1); this requires either "communication" between the two
endpoints (when one of the endpoints is configured by call/
connection controller, or simple communication of the call/
connection controller with both endpoints. Without LCAS we still
need two way communications between source and sink to coordinate
which members are used in the group and changes will not be
hitless.
Dynamic, Diverse-routes: A dynamic VCG, with the members transported
over at
least two different routes. Intent here is dynamic resizing and
resilience.
Currently support and implications similar to the "dynamic, co-
routed" and "fixed, diverse-routes" cases.
Shared Pool: Two or more VCGs between the same source and sink
who desire to share a pool of component signals between them.
Each VCG may have a dedicated set of members, and may also
obtain additional members from a "common pool" of components.
Note that at any given point in time a component signal can belong
to at most one VCG. The intent here is to allow dynamic
resizing of VCGs via the sharing of a pool of established
component signals without requiring complete circuit provisioning,
i.e., only the group membership of the component signal would change.
Currently not supported by GMPLS. Implications: a communications
mechanism between source and sink to indicate during a "change"
which group a component should now belong.
Bernstein, et al. Expires April 14, 2006 [Page 9]
Internet-Draft GMPLS, VCAT and LCAS October 2005
3. Possible Extensions to GMPLS to support additional VCAT/LCAS
scenarios
Here we look at what might be reasonable to add to GMPLS to support
the interest scenarios of Section 2 that were not currently covered.
3.1. Mechanisms for Discovery of VCAT/LCAS
Would like to get both VCAT and LCAS capability of end systems via
routing...
Would like to be able to specifically ask for LCAS capability via
signaling...
3.2. Mechanism to Support Multiple Client to End Point Mappings
This is a very important capability and it is very similar to one
that is being proposed in the end-to-end signaling for recovery I-D.
In particular the ASSOCIATION object. Note, however, since there is
a rather high probability that at some point we might use VCAT/LCAS
with GMPLS based protection we would really need an ASSOCIATION
object type specific to VCAT. Association objects are not unique and
therefore adding a new type to the Association object would make it a
good candidate to support this requirement.
3.3. Support for Component Signal Configuration Scenarios
TBD based on analysis of use of admin-status object. If the admin-
status object is sufficient we will detail its use in this
application since it is currently an optional object.
4. References
[1] Bernstein, G., Caviglia, D., and R. Rabbat, "VCAT/LCAS in a
Clamshell", To Be Published available at http://
www.grotto-networking.com/pages/VCAT-in-a-Clamshell-DRAFT.pdf,
October 2005.
[2] International Telecommunications Union, "Network node interface
for the synchronous digital hierarchy (SDH)", ITU-
T Recommendation G.707, December 2003.
[3] American National Standards Institute, "Synchronous Optical
Network (SONET) - Basic Description including Multiplex
Structure, Rates, and Formats", ANSI T1.105-2001, 2001.
[4] "Link capacity adjustment scheme (LCAS) for virtual concatenated
signals", ITU-T Recommendation G.7042, February 2004.
Bernstein, et al. Expires April 14, 2006 [Page 10]
Internet-Draft GMPLS, VCAT and LCAS October 2005
[5] "Virtual concatenation of plesiochronous digital hierarchy (PDH)
signals", ITU-T Recommendation G.7043, July 2004.
[6] "Interfaces for the Optical Transport Network (OTN)", ITU-
T Recommendation G.709, March 2003.
[7] Sklower, K., Lloyd, B., McGregor, G., Carr, D., and T.
Coradetti, "The PPP Multilink Protocol (MP)", RFC 1990,
August 1996.
[8] "Information technology - Telecommunications and information
exchange between systems - Local and metropolitan area networks
- Specific requirements - Part 3: Carrier sense multiple access
with collision detection (CSMA/CD) access method and physical
layer specifications", IEEE Standard 802.3, March 2002.
[9] Bernstein, G., Rajagopalan, B., and D. Saha, "Optical Network
Control: Archtecture, Protocols", Addison-Wesley, 2004.
Bernstein, et al. Expires April 14, 2006 [Page 11]
Internet-Draft GMPLS, VCAT and LCAS October 2005
Appendix A. Acknowledgements
The authors would like to thank Maarten Vissers for extensive reviews
and contributions to this draft.
Bernstein, et al. Expires April 14, 2006 [Page 12]
Internet-Draft GMPLS, VCAT and LCAS October 2005
Authors' Addresses
Greg Bernstein
Grotto Networking
Phone: +1 510 573 2237
Email: gregb@grotto-networking.com
Diego Caviglia
Marconi
Email: Diego.Caviglia@marconi.com
Richard Rabbat
Fujitsu
Phone: +1 408 530 4537
Email: richard@us.fujitsu.com
Bernstein, et al. Expires April 14, 2006 [Page 13]
Internet-Draft GMPLS, VCAT and LCAS October 2005
Intellectual Property Statement
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights 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; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the procedures with respect to rights in RFC documents can be
found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat 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 implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at
ietf-ipr@ietf.org.
Disclaimer of Validity
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Copyright Statement
Copyright (C) The Internet Society (2005). This document is subject
to the rights, licenses and restrictions contained in BCP 78, and
except as set forth therein, the authors retain all their rights.
Acknowledgment
Funding for the RFC Editor function is currently provided by the
Internet Society.
Bernstein, et al. Expires April 14, 2006 [Page 14]