Markdown Version | Session Recording

Session Date/Time: 29 Jul 2022 16:30

ccamp

Summary

The ccamp session included administrative updates, a reminder on IPR polling and mailing list usage, and a comprehensive review of the status of several working group documents and new drafts. Key discussions revolved around the progress and open issues for various YANG models, including microwave topology, OTN slicing, Layer 1 connectivity service, optical impairment topology, RFC 1993bis, network hardware inventory, and OTN/WSON/Flexgrid path computation. A problem statement on cloud-based services over OTN was also presented. Several drafts were deemed ready for working group adoption or last call, pending resolution of identified issues, with a strong emphasis on continuing technical discussions on the mailing list and leveraging GitHub.

Key Discussion Points

• Administrative Updates
  • The session began with load-wel and IETF policy reminders.
  • Tips for both in-person and remote participants were provided for the hybrid meeting format.
  • Chairs emphasized the importance of IPR polling responses to avoid delays in document progression.
  • Working group consensus is determined on the mailing list, even if discussed during the meeting.
• Document Status Updates
  • ISG Processing: Two documents (GMPLS for OTN beyond 100G, YANG model for flexible topology) are in ISG processing.
  • IPR Polling: Replies are still missing from 2-3 authors/contributors for the OTN Topology YANG; prompt response is requested to enable joint last call with Layer 1 Types YANG (for which all replies have been received).
  • OTN Tunnel Model: Technically stable, updated with consistent descriptions and graphics, aligned with generic TE model.
  • Transport NDI Applicability Statement: Went through WGLC, all comments addressed. Authors are reviewing and updating the document based on suggestions.
  • Client Signal YANG Model: Pending issues (resilience, port layer) are being worked on.
  • DWDM Interface LMP Extensions / Parameter YANG Model: Progressing slowly due to limited author availability.
  • Flexi-grid YANG Model: Recently adopted, but needs more discussion and review.
  • Tunnel Models for WSON and Flexi-grid: Both stable but blocked by the ietf-te model. Discussion initiated on potentially merging these two documents due to common Layer 0 attributes to reduce review effort.
• Liaisons and Communication
  • ETSI ISG liaison: Follow-up on YANG modeling for SDN SBI (microwave focus), announcing next plugfest.
  • Invitation to update IMT 2022 and beyond roadmap.
• Microwave Topology YANG Model (draft-ietf-ccamp-microwave-topology-yang)
  • Work migrated to GitHub, using IETF templates for issue tracking.
  • Minor YANG module changes, topology model figures fixed/validated, instance data added and validated.
  • Open issues: bandwidth utilization tweaks, operational mode modeling (potential update to RFC 8561).
  • ETSI liaison is using IETF YANG in plugfests, providing valuable feedback.
  • Discussion on the need for tunnel-related YANG models for microwave; chairs suggested submitting individual drafts.
• OTN Slicing Framework and Data Model (draft-ietf-ccamp-otn-slicing-yang)
  • Updated text and YANG models; NBI model shifted to augment TEAS network slicing YANG (ietf-network-slice).
  • Introduced ietf-transport-network-slice as a common layer.
  • Discussion on whether ietf-transport-network-slice should be a standalone common model or remain within the OTN slicing document.
  • Consideration of using Network Resource Partition (NRP) for OTN slicing, especially for large-scale networks or future sub-1G OTN.
  • Open issues: defining technology-specific attributes (preference for explicit attributes over opaque key-value pairs used in TEAS), and the necessity of including multiple optional MPI models in a standard draft.
• Layer 1 Connectivity Service Model (draft-ietf-ccamp-l1-csm-yang)
  • Updates: Addressed YANG Doctor review comments, provided more detailed descriptions, moved JSON example to appendix.
  • Two open issues:
    1. PM Coordination: How to coordinate performance monitoring (PM) with other documents (Math 63, ITU-T G.826, client signal PM, VPN PM, IPPM). Options presented: remove PM, Math 63 alignment only, or redo PM part.
    2. YANG Description Clarity: Clarifying constraints on protocol coding/interface functions.
• Optical Impairment Topology Model (draft-ietf-ccamp-optical-impairment-topology-yang)
  • Ongoing weekly calls. Updates to text and YANG model.
  • Started OTSi terminology alignment with G.807, clarified need for OTSi in topology model for unique identification (hierarchical approach).
  • Addressed unidirectional 3R transponder modeling, specific tests for transceiver mode.
  • Started work on protection architecture (optical protection switch). Proposed representing one OTSi with multiple associated paths for protection scenarios.
  • Open issues: completing optical protection switch modeling, supporting remote optical transponders, NETMOD/NETCONF/YANG LIR expert consultation on list usage, boundary between layers, container usage.
• RFC 1993bis (draft-ietf-ccamp-rfc1993bis)
  • Draft intends to obsolete RFC 1993 by encompassing its content with the Layer 0 type extension draft.
  • Consolidated into a single YANG module (ietf-layer-zero-type-yang).
  • Fixed several issues: added new modulation formats, corrected wrong references, refined penalty definitions (chromatic dispersion, polarization dispersion, PDL) to align with OpenConfig.
  • Clarified the meaning of central-frequency-step attribute (renamed to transceiver-tunability).
• YANG Data Model for Network Hardware Inventory (draft-ietf-ccamp-network-hardware-inventory-yang)
  • Work continued in CCAMP after no other suitable working group was found.
  • Added attributes based on RFC 8348, including common attributes (UID, name, description, location) and specific attributes for chassis/rack objects.
  • Refinements: full list of hierarchical parent identifiers for easier retrieval, placeholder for component-specific extensions, renamed model-name to part-number.
  • Discussion on location modeling: hierarchical structure vs. string attribute (OpenConfig uses string).
  • Scalability issues for large networks are acknowledged as inherent to HTTP-based transfers, not model-specific, and less impactful in daily maintenance than initial synchronization.
  • Decision to continue with the current generic component model for flexibility.
• OTN Path Computation YANG Model (draft-ietf-ccamp-otn-path-computation-yang)
  • New draft, split from WSON/Flexgrid path computation.
  • Augments TEAS path computation RPC with OTN-specific attributes from Layer 1 types, aligned with OTN tunnel model.
  • No major technical issues identified.
• WSON and Flexgrid Path Computation YANG Model (draft-ietf-ccamp-wson-flexigrid-path-computation-yang)
  • Now focused only on WSON and Flexgrid, removed OTN part.
  • Augments TEAS path computation RPC, reuses RFC 1993bis types, aligns with tunnel models.
  • Major issue: Investigating whether to merge WSON and Flexgrid computation modes into one model, given common attributes and a desire for similar merging in tunnel models. Initial assessment suggests it's possible.
• Problem Statement and Gaps for Cloud-Based Services over OTN (draft-ietf-ccamp-otn-cloud-problem-statement)
  • Motivation: Support dynamic cloud traffic over optical networks for premium services, complementing IP-based solutions.
  • Updates: Consolidated use cases, explicitly stated data plane requirements are out of scope, added gap analysis and framework description.
  • Requirements: Multi-cloud access, P2MP/MP2MP connections over OTN, dynamic bandwidth adjustment based on client traffic/QoS.
  • Identified gaps in current control plane for communicating client information between OTN PE nodes; PCEP/LS suggested as potential protocol solutions, with PCEP preferred due to perceived lack of BGP support in optical domain.
  • Proposed PCEP extensions for service information reporting (P-type report message) and service mapping configuration (P-sub-update message).
  • Discussion on the relationship with OTN slicing (could be a realization method) and Layer 1 VPNs (focus on Layer 2/3 services vs. Layer 1 services).
  • Question raised about the generality of proposed PCEP extensions (i.e., whether they are OTN-specific or common).

Decisions and Action Items

• Action Item: Authors/contributors of the OTN Topology YANG are requested to respond to the IPR polling as soon as possible to enable a joint last call with the Layer 1 Types YANG.
• Action Item: Italo and Dan are to review and update the Transport NDI Applicability document based on previous suggestions.
• Decision: The working group will discuss further on the mailing list whether the ietf-transport-network-slice module (currently in the OTN Slicing draft) should be extracted into a common, standalone document.
• Action Item: Chairs will bring the open issues for the Layer 1 Connectivity Service Model (PM coordination and YANG description clarification) to the mailing list for wider working group opinion.
• Decision: Work on the Network Hardware Inventory YANG model will continue within the ccamp working group.
• Decision: The OTN Path Computation YANG Model (draft-ietf-ccamp-otn-path-computation-yang) is deemed ready for working group adoption.
• Decision: The WSON and Flexgrid Path Computation YANG Model (draft-ietf-ccamp-wson-flexigrid-path-computation-yang) is deemed ready for working group adoption.
• Action Item: Authors are to investigate the technical feasibility and implications of merging WSON and Flexgrid modes for both the tunnel and path computation models, and then bring the options to the working group for discussion.

Next Steps

• Microwave Topology YANG Model: Address remaining open issues (bandwidth utilization, operational mode modeling) and aim for a Working Group Last Call (WGLC) by the next IETF meeting. Ensure mailing list visibility of progress.
• OTN Slicing Framework and Data Model: Address open issues (technology-specific attributes, optional MPI models), continue alignment with TEAS network slicing YANG, define OTN technology-specific models, and support both connection-based and resource-based slicing.
• Layer 1 Connectivity Service Model: Resolve the performance monitoring coordination issue and refine YANG descriptions based on mailing list feedback. Then, request a WGLC.
• Optical Impairment Topology Model: Complete the optical protection model, address remote optical transponder support, consult experts on list usage, and clarify layer boundaries. Aim for a stable version by IETF 115 for YANG Doctor review.
• RFC 1993bis: Reconcile introductory text from RFC 1993 and the Layer 0 type extension draft, complete Appendix A (listing changes), and fix remaining GitHub issues.
• Network Hardware Inventory YANG Model: Prepare for a second working group adoption call, solicit input from other technologies (IP, microwave), and continue addressing open issues.
• OTN Path Computation YANG Model: Solicit feedback, ensure alignment with OTN tunnel model, and complete security/manageability considerations.
• WSON and Flexgrid Path Computation YANG Model: Obtain feedback, make editorial updates, evaluate the option of merging WSON and Flexgrid modes, ensure alignment with related tunnel models and RFC 1993bis, and complete security/manageability considerations.
• Problem Statement and Gaps for Cloud-Based Services over OTN: Integrate the framework section, prepare the draft for working group adoption, and solicit further comments to initiate discussions on potential protocol extensions.