NOTE: This charter is a snapshot of the 50th IETF Meeting in Minneapolis, Minnesota. It may now be out-of-date. Last Modified: 14-Mar-01
James Luciani <firstname.lastname@example.org>
Daniel Awduche <email@example.com>
Fred Baker <firstname.lastname@example.org>
Fred Baker <email@example.com>
David Oran <firstname.lastname@example.org>
Thomas Narten <email@example.com>
To Subscribe: http://lists.bell-labs.com/mailman/listinfo/ip-optical
Note: this WG is one of a set of WGs that comprise the 'sub-IP' pseudo-area that the IESG recently created. All of the sub-IP WGs are temporarily being housed in the General Area until the IESG decides on a final management structure for managing these groups. The Area Directors for the following areas acting as a group will be the Area Advisors: INT, OPS, RTG and TSV. The name(s) listed above under "Technical Advisor(s)" are the the responsible Area Directors for the working group.
The advent of switched multi-channel (e.g., WDM, DWDM, OTDM) optical networks using OXCs (Optical Cross-connects) and other optical switching elements presents many new opportunities for improving the performance of IP networks and supporting faster and more flexible provisioning of IP services. However, much needs to be specified before interoperable products based on these technologies can be deployed in service provider networks. The work needed in this area includes:
- Document the use of existing framing methods for IP over optical dataplane and control channels, and as necessary specify additional framing methods.
- Identify and document the characteristics of the optical transport network that are important for selecting paths for optical channels, setting-up optical channels, and tearing-down optical channels.
- Document the applications of the common control and measurement protocols to the technology-dependent aspects of optical path setup, teardown, and maintenance of optical channels across networks with optical components.
- Document the requirements for control of optical networks by elements outside the optical network itself.
- Document the applicability of IP-based protocols for the controlled dissemination of optical network topology, metric, and constraint information. Such information can be used for inventory management, path selection, and other purposes. The information to be exchanged should accommodate both all-optical and optical-electrical-optical switching technologies.
- If a need is identified to develop new protocols or make incompatible modifications to existing protocols (e.g., routing and signaling protocols) to accomplish the above goals, then a recharter must first be approved before undertaking such work.
The IP over Optical WG will coordinate with relevant working groups within the IETF to leverage existing work. The WG may also generate requirements for other IETF WGs as needed. Additionally, the WG may collaborate with other standards bodies and interoperability forums engaged in IP over optical activities (e.g., including ITU-T) to share information, minimize duplication of effort, and coordinate activities in order to promote interoperability and serve the best interest of the industry.
Produce Framework document with adequate motivation and detailed description of the context and solution space for IP/optical.
Submit I-D on 'carrier optical services' requirements for IP/Optical networks.
Submit I-D on the unique features of optical networks that are relevant for path computation and signaling in IP/Optical contexts.
Submit I-D on the Traffic Engineering issues that are specific to IP/Optical networks, which take into account the peculiar characteristics of the operating environment, including aspects of protection/restoration unique to IP/optical.
Submit I-D on the requirements for distribution of optical topology state information using the common control and measurement protocols.
Submit I-D on the requirements for signaling in IP/Optical networks using IP signaling protocols.
Submit applicability statements describing the use of IP based protocols for signaling and dissemination of network topology state information in IP/Optical networks.
Document existing encapsulation schemes for IP/optical data plane and control plane channels.
Submit Framework draft to IESG as a informational RFC.
1. Temporary sub-IP area has been created.
2. The scope is not entirely clear at this point.
3. The temp. area will be short-lived, 2 years, say. 2 ADs temporarily assigned. Scott and Bert.
4. Need to determine where all the IDs belong (among all WGs in the new area).
5. Reevaluate organization after the London meeting.
Jim Luciani, Chair:
2. More than 20 slot requests were received, but couldn't accommodate all.
3. IPO is now a WG.
4. Agenda description.
Jim Luciani, presenting draft-chiu-strand-unique-olcp-02.txt
1. Described the delta from the previous version.
2. Ongoing work: crankback, limited wavelength conversion, heterogeneous fiber types, security.
3. Dan: The doc. will be made the basis of a design team document. The doc. will also be made a WG document.
Bala: Framework doc.
1. Not much change from last version.
2. Bala said that the draft focusses on protocol and control architectures but may need to expand scope based on the IPO scope.
3. Draft will be made a wg document
4. Question: will the draft include pointers to other drafts being included in IPO?
Jim: The draft should cover these issues. Whether this doc alone, or another new one will have pointers it TBD.
Ayan Banerjee: draft-banerjee-routing-impairments-00.txt
1. Optical impairment parameters have been narrowed down to three main parameters. These are used for determining acceptable paths.
2. Described the parameters in some detail.
Dan: this draft will be consolidated with Angela Chiu's draft. Ask those interested to contribute to the design team effort.
3. Greg Bernstien: Where and when will the measurements be made?
Ayan: Some will be known based on the equipment. Others may be statistically determined.
4. Eve Verma: If you don't know what's being carried on the link, how can do you know the SNR is sufficient?
Jim: Need to take the worst case values?
Ayan: SNR can be a table of values based on bit rate.
Jim: Network operator can determine SNR values based on media types, by apiori measurements.
Eve: The intent for this group to do detailed optical span engineering?
Dan: No, the intent is to take into account link charac. and impairments when computing paths.
Eve: Is the proposal to do the computations and propagation in real time?
Dan: There are different models. We are looking for engineering solutions to take into account accummulative impairments in path computation.
Greg: Does the draft require measurements to be taken in a particular way? This shouldn't be the case.
Jim: That's right. This WG should be specifying the requirements, the CCAMP group will be the one to specify how the measurements are taken and carried.
Unknown: How does this all relate to IP?
Dan: Since IP clients are important for optical network, IETF has an interest in this topic. In order to develop a control plane for optical networks, it is necessary to understand the constraints.
Greg Bernstien: Optical inter-domain routing
1. Aimed at requirements only in the specific case of optical internetworks.
2. Motivation for inter-domain routing described.
3. Differences between IP and optical domains for inter-domain routing described.
4. Yanguang: Diversity across domains is typically of no interest to clients. Diversity is an intra-domain issue.
Jim: There are customers who ask for end-to-end diversity. My question is how to actually enforce this.
Unknown: When it comes to geographical diversity, the federal govt. has specific requirements on diversity. Will this be covered?
Greg: The idea may be to use even approximate information.
Jim: If I'm a carrier, I'm not going to tell you my internal topology.
Unknown: The comparison should be between MPLS and optical routing, not IP and optical domain routing.
Nasir Ghani: Architectural framework for automatic protection provisioning in Dynamic optical rings.
Nasir: Wants ring network considerations taken into account in protocols for provisioning and restoration.
W.r.t restoration, either use MPLS schemes or a new APS protocol
Dimitri Papadimitriou: draft-papadimitriou-optical-rings-00.txt
Model: Emulate optical ring over a mesh topology
Rob: Need a clear determination that carriers and SPs need to pursue this type of work. Can be in IPO or TE WG. There is also the IPORPR work and how to combine.
Jim: Need to really motivate the need for rings over mesh. This info needs to be in the draft.
Unknown: Same type of work is well-underway in ITU. Is there any idea of collaboration or coordination with ITU? ***(See Notes At End)
Dimitry: The ITU work is still under discussion, and the extension for optical has not been taken up.
Greg: The tricky part is to make sure that there is interop. between GMPLS and routing protocols.
Yong Xue: The carriers are interested in this work. The concern is also interworking between the sonet and optical ring control.
Concept: To support the "light-tree" concept (point-to-multipoint).
Described the concept and constraints on optical multicast.
Jim: The draft is getting into solutions rather than requirements.
Jim: Some parts are important, e.g., how the splitters may have to be configured for specific multicast. Taking into account loss, etc.
Vijay: The draft seems to be addressing issues that noone is interested in at this point.
Jim: there is actually interest in this particular problem.
Yong Xue: Carrier Optical Services Framework
Based on the work at OIF.
Welcome other carriers to participate.
Proposal: Accept document as a WG document.
Unknown: Not that much on security in the document.
Yong Xue: Security is a big issue. It's been spread around in the document.
Jim: Going to consult the new area ADs to see if UNI etc (and this doc) will be part of this WG.
Ishimatsu: Security requirements for lightpath services
Proposed adopting the draft as the WG doc.
Dan: proposed consolidating this doc with the carrier services document.
Eric Mannie: ITU is accepting requirements. To what extent will the accepted requirements be satisfied by CCAMP?
Dan: We're first trying to understand the requirements. Not all requirements may be satisfied in protocol work.
Dan: We agree that this doc. will be consolidated with the carrier services framework doc.
Osama: ASON and related protocols
Osama: The purpose is informational.
Osama: ITU defines architectural models. The idea is to reuse IETF protocols for signaling and routing.
Yong Xue: Two parallel stds. activities going on, ITU and IETF. The concern is whether there will be two sets of inconsistent standards. How do we make sure that the two parallel activities will not conflict with each other?
Jim: We need to coordinate with other bodies closely, specifically, ITU. Given the number of people who attend both, coordination is definitely possible.
Closing remarks: Dan Awduche.
Dan: Consolidate Chiu and banerjee drafts to form the basis for a design team
Convert framework draft into WG doc.
Consolidate the carrier services and security draft into one.
No decision on other drafts yet.
***The following was a point made on the list by Zhi-Wei Lin: "I would like to point out that the ITU and T1X1.5 has taken up optical extensions for protection for a while now. Actually it was taken up since 1999. We've had quite a bit of progress, with G.872 describing the high level protection requirements, G.798 describing the components necessary to support protection, and G.841 describing the protocols for the rings."
A New Area
Automatic Switched Optical Networks (ASON) and its Related Protocols
Carrier Optical Services Framework and Associated UNI Requirements
Unique Features and Requirements for The Optical Layer Control Plane
Security Requirements for Lightpath Services
Optical Rings and Hybrid Mesh Rings Optical Networks
Optical Multicast - Framework
Optical Inter-domain routing
Architectural Framework For Automatic Protection Provisioning in Dynamic Optical Rings
Impairment Constraints for Routing in All Optical Networks