PCEP for WSON: Path Computation Element Communication Protocol (PCEP) Requirements for Wavelength Switched Optical Network (WSON) Routing and Wavelength Assignment
What a path computation element has to be told to place an optical path and pick its wavelength.
Lee, Y., Ed., Bernstein, G., Ed., Martensson, J., Takeda, T., Tsuritani, T., and O. Gonzalez de Dios, "Path Computation Element Communication Protocol (PCEP) Requirements for Wavelength Switched Optical Network (WSON) Routing and Wavelength Assignment", RFC 7449, DOI 10.17487/RFC7449, February 2015, <https://www.rfc-editor.org/info/rfc7449>.
@misc{rfc7449,
series = {Request for Comments},
number = 7449,
howpublished = {RFC 7449},
publisher = {RFC Editor},
doi = {10.17487/RFC7449},
url = {https://www.rfc-editor.org/info/rfc7449},
author = {Y. Lee and G. Bernstein and J. Martensson and T. Takeda and T. Tsuritani and O. Gonzalez de Dios},
title = {{Path Computation Element Communication Protocol (PCEP) Requirements for Wavelength Switched Optical Network (WSON) Routing and Wavelength Assignment}},
pagetotal = 12,
year = 2015,
month = feb,
} RFC text
Contents
Internet Engineering Task Force (IETF) Y. Lee, Ed.
Request for Comments: 7449 Huawei
Category: Informational G. Bernstein, Ed.
ISSN: 2070-1721 Grotto Networking
J. Martensson
Acreo
T. Takeda
NTT
T. Tsuritani
KDDI
O. Gonzalez de Dios
Telefonica
February 2015
Path Computation Element Communication Protocol (PCEP) Requirements
for Wavelength Switched Optical Network (WSON)
Routing and Wavelength Assignment
Abstract
This memo provides application-specific requirements for the Path
Computation Element Communication Protocol (PCEP) for the support of
Wavelength Switched Optical Networks (WSONs). Lightpath provisioning
in WSONs requires a Routing and Wavelength Assignment (RWA) process.
From a path computation perspective, wavelength assignment is the
process of determining which wavelength can be used on each hop of a
path and forms an additional routing constraint to optical light path
computation. Requirements for PCEP extensions in support of optical
impairments will be addressed in a separate document.
Status of This Memo
This document is not an Internet Standards Track specification; it is
published for informational purposes.
This document is a product of the Internet Engineering Task Force
(IETF). It represents the consensus of the IETF community. It has
received public review and has been approved for publication by the
Internet Engineering Steering Group (IESG). Not all documents
approved by the IESG are a candidate for any level of Internet
Standard; see Section 2 of RFC 5741.
Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
http://www.rfc-editor.org/info/rfc7449.
Copyright Notice
Copyright (c) 2015 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Table of Contents
1. Introduction ....................................................3
2. WSON RWA Processes and Architecture .............................4
3. Requirements ....................................................5
3.1. Path Computation Type Option ...............................5
3.2. RWA Processing .............................................6
3.3. Bulk RWA Path Request/Reply ................................6
3.4. RWA Path Reoptimization Request/Reply ......................7
3.5. Wavelength Range Constraint ................................7
3.6. Wavelength Assignment Preference ...........................7
3.7. Signal-Processing Capability Restriction ...................8
4. Manageability Considerations ....................................8
4.1. Control of Function and Policy .............................8
4.2. Information and Data Models (e.g., MIB Module) .............9
4.3. Liveness Detection and Monitoring ..........................9
4.4. Verifying Correct Operation ................................9
4.5. Requirements on Other Protocols and Functional Components ..9
4.6. Impact on Network Operation ................................9
5. Security Considerations .........................................9
6. References .....................................................10
6.1. Normative References ......................................10
6.2. Informative References ....................................10
Acknowledgments....................................................11
Authors' Addresses.................................................11
1. Introduction
[RFC4655] defines the PCE-based architecture and explains how a Path
Computation Element (PCE) may compute Label Switched Paths (LSPs) in
networks controlled by Multiprotocol Label Switching Traffic
Engineering (MPLS-TE) and Generalized MPLS (GMPLS) at the request of
Path Computation Clients (PCCs). A PCC is shown to be any network
component that makes such a request and may be, for instance, an
optical switching element within a Wavelength Division Multiplexing
(WDM) network. The PCE itself can be located anywhere within the
network; it may be within an optical switching element, a Network
Management System (NMS), or an Operational Support System (OSS), or
it may be an independent network server.
The Path Computation Element Communication Protocol (PCEP) is the
communication protocol used between a PCC and PCE; it may also be
used between cooperating PCEs. [RFC4657] sets out the common
protocol requirements for PCEP. Additional application-specific
requirements for PCEP are deferred to separate documents.
This document provides a set of application-specific PCEP
requirements for support of path computation in Wavelength Switched
Optical Networks (WSONs). WSON refers to WDM-based optical networks
in which switching is performed selectively based on the wavelength
of an optical signal.
The path in WSON is referred to as a lightpath. A lightpath may span
multiple fiber links, and the path should be assigned a wavelength
for each link.
A transparent optical network is made up of optical devices that can
switch but not convert from one wavelength to another. In a
transparent optical network, a lightpath operates on the same
wavelength across all fiber links that it traverses. In such cases,
the lightpath is said to satisfy the wavelength-continuity
constraint. Two lightpaths that share a common fiber link cannot be
assigned the same wavelength. To do otherwise would result in both
signals interfering with each other. Note that advanced additional
multiplexing techniques such as polarization-based multiplexing are
not addressed in this document since the physical-layer aspects are
not currently standardized. Therefore, assigning the proper
wavelength on a lightpath is an essential requirement in the optical
path computation process.
When a switching node has the ability to perform wavelength
conversion, the wavelength-continuity constraint can be relaxed, and
a lightpath may use different wavelengths on different links along
its path from origin to destination. It is, however, to be noted
that wavelength converters may be limited for cost reasons, while the
number of WDM channels that can be supported in a fiber is also
limited. As a WSON can be composed of network nodes that cannot
perform wavelength conversion, nodes with limited wavelength
conversion, and nodes with full wavelength conversion abilities,
wavelength assignment is an additional routing constraint to be
considered in all lightpath computations.
In this document, we first review the processes for Routing and
Wavelength Assignment (RWA) used when wavelength continuity
constraints are present and then specify requirements for PCEP to
support RWA. Requirements for optical impairments will be addressed
in a separate document.
The remainder of this document uses terminology from [RFC4655].
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 [RFC2119].
2. WSON RWA Processes and Architecture
In [RFC6163], three alternative process architectures were given for
performing routing and wavelength assignment. These are shown
schematically in Figure 1, where R stands for Routing, WA for
Wavelength Assignment, and DWA for Distributed Wavelength Assignment.
+-------------------+
| +-------+ +--+ | +-------+ +--+ +-------+ +---+
| | R | |WA| | | R |--->|WA| | R |--->|DWA|
| +-------+ +--+ | +-------+ +--+ +-------+ +---+
| Combined | Separate Processes Separate Processes
| Process | WA performed in a
+-------------------+ distributed manner
(a) (b) (b')
Figure 1: RWA Process Alternatives
These alternatives have the following properties and impact on PCEP
requirements in this document.
(a) Combined Process (R&WA)
Path selection and wavelength assignment are performed as a
single process. The requirements for PCC-PCE interaction with a
PCE implementing such a combined RWA process are addressed in
this document.
(b) Routing Separate from Wavelength Assignment (R+WA)
The routing process furnishes one or more potential paths to the
wavelength assignment process that then performs final path
selection and wavelength assignment. The requirements for PCE-
PCE interaction with one PCE implementing the routing process
and another implementing the wavelength assignment process are
not addressed in this document.
(b') Routing and Distributed Wavelength Assignment (R+DWA)
A standard path computation (unaware of detailed wavelength
availability) takes place, and then wavelength assignment is
performed along this path in a distributed manner via signaling
(RSVP-TE). This alternative is a particular case of R+WA and
should be covered by GMPLS PCEP extensions; it does not present
new WSON-specific requirements.
The various process architectures for implementing RWA have been
reviewed above. Figure 2 shows one typical PCE-based implementation,
which is referred to as the Combined Process (R&WA). With this
architecture, the two processes of routing and wavelength assignment
are accessed via a single PCE. This architecture is the base
architecture from which the requirements are specified in this
document.
+----------------------------+
+-----+ | +-------+ +--+ |
| | | |Routing| |WA| |
| PCC |<----->| +-------+ +--+ |
| | | |
+-----+ | PCE |
+----------------------------+
Figure 2: Combined Process (R&WA) Architecture
3. Requirements
The requirements for the PCC-to-PCE interface of Figure 2 are
specified in this section.
3.1. Path Computation Type Option
A PCEP request MAY include the path computation type. This can be:
(a) Both RWA, or
(b) Routing only.
This requirement is needed to differentiate between the currently
supported routing with distributed wavelength assignment option and
combined RWA. For the distributed wavelength assignment option,
wavelength assignment will be performed at each node of the route.
3.2. RWA Processing
As discussed in Section 2, various RWA processing options should be
supported in a PCEP request/reply.
(a) When the request is an RWA path computation type, the request
MUST further include the wavelength assignment options. At
minimum, the following options should be supported:
(i) Explicit Label Control (ELC) [RFC3473]
(ii) A set of recommended labels for each hop. The PCC can
select the label based on local policy.
Note that option (ii) may also be used in R+WA or R+DWA.
(b) In case of an RWA computation type, the response MUST include
the wavelength(s) assigned to the path and an indication of
which label assignment option has been applied (ELC or label
set).
(c) In the case where a valid path is not found, the response MUST
include why the path is not found (e.g., network disconnected,
wavelength not found, both, etc.). Note that 'wavelength not
found' may include several sub-cases such as wavelength
continuity not met, unsupported FEC/Modulation type, etc.
3.3. Bulk RWA Path Request/Reply
Sending simultaneous path requests for "routing only" computation is
supported by the PCEP specification [RFC5440]. To remain consistent,
the following requirements are added.
(a) A PCEP request MUST be able to specify an option for bulk RWA
path requests. A bulk path request provides an ability to
request a number of simultaneous RWA path requests.
(b) The PCEP response MUST include the path and the assigned
wavelength for each RWA path request specified in the original
bulk request.
3.4. RWA Path Reoptimization Request/Reply
This section provides a number of requirements concerning RWA path
reoptimization processing in PCEP.
(a) For a reoptimization request, the request MUST provide both the
path and current wavelength to be reoptimized and MAY include
the following options:
(i) Reoptimize the path keeping the same wavelength(s)
(ii) Reoptimize wavelength(s) keeping the same path
(iii) Reoptimize allowing both the wavelength and the path to
change
(b) The corresponding response to the reoptimized request MUST
provide the reoptimized path and wavelengths even when the
request asked for the path or the wavelength to remain
unchanged.
(c) In the case that the new path is not found, the response MUST
include why the path is not found (e.g., network disconnected,
wavelength not found, both, etc.). Note that 'wavelength not
found' may include several sub-cases such as wavelength
continuity not met, unsupported FEC/Modulation type, etc.
3.5. Wavelength Range Constraint
For any RWA computation type request, the requester (PCC) MUST be
allowed to specify a restriction on the wavelengths to be used. The
requester MAY use this option to restrict the assigned wavelength for
explicit labels or label sets. This restriction may, for example,
come from the tuning ability of a laser transmitter, any optical
element, or a policy-based restriction.
Note that the requester (e.g., PCC) is not required to furnish any
range restrictions.
3.6. Wavelength Assignment Preference
In a network with wavelength conversion capabilities (e.g., sparse 3R
regenerators), a request SHOULD be able to indicate whether a single,
continuous wavelength should be allocated or not. In other words,
the requesting PCC SHOULD be able to specify the precedence of
wavelength continuity even if wavelength conversion is available.
(a) An RWA computation type request MAY include the requester
preference for random assignment, descending order, ascending
order, etc. A response SHOULD follow the requester preference
unless it conflicts with the operator's policy.
(b) A request for two or more paths MUST allow the requester to
include an option constraining the paths to have the same
wavelength(s) assigned. This is useful in the case of
protection with a single transponder (e.g., 1+1 link disjoint
paths).
3.7. Signal-Processing Capability Restriction
Signal-processing compatibility is an important constraint for
optical path computation. The signal type for an end-to-end optical
path must match at the source and at the destination.
The PCC MUST be allowed to specify the signal type at the endpoints
(i.e., at the source and at the destination). The following signal-
processing capabilities should be supported at a minimum:
o Modulation Type List
o FEC Type List
The PCC MUST also be allowed to state whether transit modification is
acceptable for the above signal-processing capabilities.
4. Manageability Considerations
Manageability of WSON RWA with PCE must address the following
considerations.
4.1. Control of Function and Policy
In addition to the parameters already listed in Section 8.1 of
[RFC5440], a PCEP implementation SHOULD allow configuring the
following PCEP session parameters on a PCC:
o The ability to send a WSON RWA request.
In addition to the parameters already listed in Section 8.1 of
[RFC5440], a PCEP implementation SHOULD allow configuring the
following PCEP session parameters on a PCE:
o The support for WSON RWA.
o The maximum number of bulk path requests associated with WSON RWA
per request message.
These parameters may be configured as default parameters for any PCEP
session the PCEP speaker participates in, or may apply to a specific
session with a given PCEP peer or a specific group of sessions with a
specific group of PCEP peers.
4.2. Information and Data Models
As this document only concerns the requirements to support WSON RWA,
no additional MIB module is defined in this document. However, the
corresponding solution document will list the information that should
be added to the PCE MIB module defined in [RFC7420].
4.3. Liveness Detection and Monitoring
This document does not define any new mechanisms that imply any new
liveness detection and monitoring requirements in addition to those
already listed in Section 8.3 of [RFC5440].
4.4. Verifying Correct Operation
This document does not define any new mechanisms that imply any new
verification requirements in addition to those already listed in
Section 8.4 of [RFC5440]
4.5. Requirements on Other Protocols and Functional Components
If PCE discovery mechanisms ([RFC5089] and [RFC5088]) were to be
extended for technology-specific capabilities, advertising WSON RWA
path computation capability should be considered.
4.6. Impact on Network Operation
This document does not define any new mechanisms that imply any new
network operation requirements in addition to those already listed in
Section 8.6 of [RFC5440].
5. Security Considerations
This document has no requirement for a change to the security models
within PCEP [RFC5440]. However, the additional information
distributed in order to address the RWA problem represents a
disclosure of network capabilities that an operator may wish to keep
private. Consideration should be given to securing this information.
Solutions that address the requirements in this document need to
verify that existing PCEP security mechanisms adequately protect the
additional network capabilities and must include new mechanisms as
necessary.
6. References
6.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997,
<http://www.rfc-editor.org/info/rfc2119>.
[RFC4655] Farrel, A., Vasseur, J.-P., and J. Ash, "A Path
Computation Element (PCE)-Based Architecture", RFC 4655,
August 2006, <http://www.rfc-editor.org/info/rfc4655>.
[RFC5440] Vasseur, JP., Ed., and JL. Le Roux, Ed., "Path Computation
Element (PCE) Communication Protocol (PCEP)", RFC 5440,
March 2009, <http://www.rfc-editor.org/info/rfc5440>.
6.2. Informative References
[RFC3473] Berger, L., Ed., "Generalized Multi-Protocol Label
Switching (GMPLS) Signaling Resource ReserVation Protocol-
Traffic Engineering (RSVP-TE) Extensions", RFC 3473,
January 2003, <http://www.rfc-editor.org/info/rfc3473>.
[RFC4657] Ash, J., Ed., and J. Le Roux, Ed., "Path Computation
Element (PCE) Communication Protocol Generic
Requirements", RFC 4657, September 2006,
<http://www.rfc-editor.org/info/rfc4657>.
[RFC5088] Le Roux, JL., Ed., Vasseur, JP., Ed., Ikejiri, Y., and R.
Zhang, "OSPF Protocol Extensions for Path Computation
Element (PCE) Discovery", RFC 5088, January 2008,
<http://www.rfc-editor.org/info/rfc5088>.
[RFC5089] Le Roux, JL., Ed., Vasseur, JP., Ed., Ikejiri, Y., and R.
Zhang, "IS-IS Protocol Extensions for Path Computation
Element (PCE) Discovery", RFC 5089, January 2008,
<http://www.rfc-editor.org/info/rfc5089>.
[RFC6163] Lee, Y., Ed., Bernstein, G., Ed., and W. Imajuku,
"Framework for GMPLS and Path Computation Element (PCE)
Control of Wavelength Switched Optical Networks (WSONs)",
RFC 6163, April 2011,
<http://www.rfc-editor.org/info/rfc6163>.
[RFC7420] Koushik, A., Stephan, E., Zhao, Q., King, D., and J.
Hardwick, "Path Computation Element Communication Protocol
(PCEP) Management Information Base (MIB) Module", RFC
7420, December 2014,
<http://www.rfc-editor.org/info/rfc7420>.
Acknowledgments
The authors would like to thank Adrian Farrel, Cycil Margaria, and
Ramon Casellas for many helpful comments that greatly improved the
content of this document.
Authors' Addresses
Young Lee (editor)
Huawei Technologies
5340 Legacy Drive, Building 3
Plano, TX 75245
United States
Phone: (469) 277-5838
EMail: leeyoung@huawei.com
Greg Bernstein (editor)
Grotto Networking
Fremont, CA
United States
Phone: (510) 573-2237
EMail: gregb@grotto-networking.com
Jonas Martensson
Acreo
Isafjordsgatan 22
164 40 Kista
Sweden
EMail: Jonas.Martensson@acreo.se
Tomonori Takeda
NTT Corporation
3-9-11, Midori-Cho
Musashino-Shi, Tokyo 180-8585
Japan
EMail: tomonori.takeda@ntt.com
Takehiro Tsuritani
KDDI R&D Laboratories, Inc.
2-1-15 Ohara Kamifukuoka Saitama, 356-8502
Japan
Phone: +81-49-278-7806
EMail: tsuri@kddilabs.jp
Oscar Gonzalez de Dios
Telefonica
Distrito Telefonica, ed. Sur 3, Pta 3, Ronda de la Comunicacion
Madrid, 28050
Spain
Phone: +34 913129647
EMail: oscar.gonzalezdedios@telefonica.com
ISO Plain Language
On this page
§1What a Wavelength Switched Optical Network Is
The path computation element architecture describes a component that computes label switched paths for packet networks at the request of a client. The client may be any network component, and in the case of an optical network it may be an optical switching element inside a wavelength division multiplexing network. The element itself may sit anywhere: inside an optical switch, in a network management system or operational support system, or on its own server. The communication protocol between client and element may also be used between cooperating elements, and the common requirements for that protocol are set out separately, with application-specific requirements left to other documents. This one covers the case of path computation in wavelength switched optical networks, where the switching is done selectively according to the wavelength of the optical signal.
The path through such a network is called a lightpath. It may cross several fibre links, and a wavelength should be assigned for each of them. A transparent optical network is built from devices that can switch but cannot convert one wavelength into another, so a lightpath there runs on the same wavelength across every fibre it crosses, which is called the wavelength-continuity constraint. Two lightpaths that share a fibre link cannot be given the same wavelength, because the signals would interfere with each other. The document leaves out advanced multiplexing such as polarization-based multiplexing, because the physical layer aspects are not standardised, which makes choosing the right wavelength an essential part of optical path computation.
Where a node can convert wavelengths, the continuity constraint can be relaxed and a lightpath may use different wavelengths on different links. Converters may be limited for cost reasons, and the number of wavelengths a fibre can carry is also limited, and a network may mix nodes with no conversion, limited conversion and full conversion. Wavelength assignment is therefore an extra constraint that every lightpath computation has to take into account.
An optical network sends light down fibres, and each fibre carries several wavelengths at once, each on its own signal. A lightpath is a route through the network, and it needs one wavelength per fibre. Where the equipment cannot change the wavelength along the way, the same wavelength must be used on every fibre, and no two lightpaths on the same fibre may use the same wavelength.
§2Three Ways to Route and Assign Wavelengths
Three alternative process architectures for routing and wavelength assignment were set out earlier, and they shape the requirements in this document. In the combined process, path selection and wavelength assignment are done together as a single process. In routing separate from wavelength assignment, the routing process produces one or more candidate paths and the wavelength assignment process then makes the final choice of path and of wavelengths; the requirements for a client talking to one element that routes and another that assigns wavelengths are not addressed here. In routing with distributed wavelength assignment, a standard path computation happens without any knowledge of wavelength availability, and the wavelengths are then assigned along that path in a distributed way through signalling. That last case is a particular form of the separate approach, is expected to be covered by the Generalized MPLS extensions of the communication protocol, and raises no new requirements specific to optical networks.
The requirements in this document are written against the combined process. In that architecture, one element offers both functions, the routing and the wavelength assignment, and the client talks to it over the path computation protocol.
There are three ways to organise the work. One element can choose the route and the wavelengths together. Two elements can split the job, one choosing routes and one choosing wavelengths. Or a route can be chosen without any knowledge of wavelengths and the wavelengths assigned later as the path is set up. The document takes the first of these as its model.
| Process | What it does | Covered here |
|---|---|---|
| Combined (R&WA) | Path selection and wavelength assignment are one process. | Yes |
| Separate (R+WA) | Routing gives candidate paths; wavelength assignment makes the final choice. | No |
| Routing with distributed assignment (R+DWA) | A normal path computation, then wavelengths assigned along the path by signalling. | No, expected to be covered by the GMPLS extensions |
§3What the Request and the Reply Must Carry
A request may carry the path computation type, which is either routing with wavelength assignment or routing only. The distinction is needed so that the combined process can be told apart from the distributed wavelength assignment option, where wavelengths are assigned at each node along the route as it is set up.
When the request is for routing with wavelength assignment, it must also carry the wavelength assignment options, and at least two should be supported: explicit label control, and a set of recommended labels for each hop, from which the client can pick one according to local policy. The second of these can also be used in the other two process models. The reply to such a request must include the wavelengths assigned to the path and say which of the two label assignment options was applied. Where no valid path is found, the reply must say why, for example that the network is disconnected or that no wavelength was found, and the wavelength case can have sub-cases such as the continuity constraint not being met or an unsupported forward error correction or modulation type.
Requests for routing only can already be sent in bulk by the protocol, and the same is required for routing with wavelength assignment: a request must be able to ask for a number of simultaneous computations, and the reply must give the path and the assigned wavelength for each of them.
Reoptimization is also covered. A reoptimization request must provide both the path and the current wavelength, and may say whether the wavelength should be kept while the path may change, whether the path should be kept while the wavelength may change, or whether both may change. The reply must give the reoptimized path and wavelengths even when the request asked for one of them to stay the same, and if no new path is found it must say why, in the same terms as before.
The request has to say whether it wants a route only or a route with wavelengths. If it wants wavelengths, it must also say how the wavelengths should be chosen: either fixed by the element, or as a list of suggestions the client can pick from. The reply has to state the wavelengths it used and say which method it used, and if it could not find a path it has to say whether the problem was the network or the wavelengths.
| Option | What may change | What must not |
|---|---|---|
| Keep the wavelength | The path | The wavelength |
| Keep the path | The wavelength | The path |
| Allow both to change | The path and the wavelength | Nothing |
- The wavelengths assignedWhich wavelength was chosen for each link of the path.
- The method usedWhether the wavelengths came from explicit label control or from a set of suggestions.
- Why nothing could be assignedThe network is disconnected, no wavelength was found, or both.
- Which wavelength failedContinuity could not be met, or the signal type was not supported.
- Every path in a bulk requestThe path and the assigned wavelength for each request that was sent.
- The reoptimized resultThe path and the wavelengths, even if one of them was asked to stay the same.
§3.5Wavelength Ranges, Preferences and Signal Types
The client must be allowed to restrict which wavelengths may be used. The restriction might come from the tuning range of a laser transmitter, from the capability of some other optical element, or from a policy decision. The client is not obliged to supply any restriction.
Where the network has wavelength conversion, for example at sparse regenerators, the request should be able to say whether a single continuous wavelength is wanted anyway, so that the client can set the precedence of the continuity constraint even where it could be relaxed. The request may also state a preference for how the wavelengths are picked, such as random, descending or ascending order, and the reply should follow that preference unless it conflicts with the operator's policy. A request for two or more paths must allow the client to constrain them to the same wavelengths, which is what protection with a single transponder needs, such as a pair of link-disjoint paths for 1+1 protection.
Signal processing is the other constraint that matters. The signal type of an end-to-end optical path must match at the source and the destination, so the client must be able to state the signal type at the endpoints. At a minimum, a modulation type list and a forward error correction type list should be supported, and the client must also be able to say whether transit modification of those capabilities is acceptable.
Three more things shape the answer. The client can rule out wavelengths that its hardware cannot produce or that policy forbids. It can ask for one continuous wavelength even where the network could change it, and can express a preference for how the wavelengths are picked. And it must be able to say what kind of signal it needs at each end, because the signal has to be the same at both ends of the path.
§4Running It in Practice
Manageability has to cover a few practical points. A client should be able to configure the ability to send a request for routing with wavelength assignment. On the element's side, the session parameters should include support for that kind of request and the maximum number of bulk path requests that may be associated with it in one request message. These parameters may be set as defaults for any session the speaker takes part in, or for a specific session with a given peer or a specific group of sessions with a given group of peers.
The document defines no new management information module, because it only covers requirements; the matching solution document will list what should be added to the path computation element's management module. It defines no new liveness detection or monitoring mechanisms beyond those already in the protocol, and no new verification requirements beyond those already there. If the discovery mechanisms were extended for technology-specific capabilities, advertising the ability to compute optical paths should be considered. And no new network operation requirements are implied.
The operational side is mostly left to the existing protocol. The document adds two things an operator can configure: whether a client may send these requests at all, and how many such requests may be sent in one message. It also notes that if the protocol's discovery mechanism is ever extended to advertise what an element can do, this ability should be one of the things advertised.
§5Security Considerations
The document requires no change to the security model of the path computation protocol. It does note, however, that the extra information exchanged to solve the wavelength assignment problem discloses facts about the network's capabilities that an operator may want to keep private, and that securing this information should be considered. A solution that meets these requirements has to check that the existing security mechanisms of the protocol adequately protect the additional network capabilities, and must include new mechanisms where they do not.
The document does not change the protocol's security rules, but it points out that telling the element what wavelengths are available, and what the equipment can do, reveals more about the network than before. An operator who wants that kept private needs the exchange to be protected.