Session Date/Time: 26 Jul 2024 16:30

bmwg

Summary

The Benchmarking Methodology Working Group (BMWG) session focused on updates to existing drafts, discussion of new proposals, and a presentation on benchmarking IoT devices. Key topics included multiple loss ratio search, a YANG data model for network tester management, benchmarking network performance in container infrastructure, benchmarking segmented audio, power in network devices, SRv6 service benchmarking, source address validation, and benchmarking IoT devices. The group discussed the stability of existing drafts and the need for more reviews.

Key Discussion Points

Multiple Loss Ratio (MLR) Search Draft: Authors addressed previous comments and believe the draft is ready for last call. Discussion focused on the architecture (manager, controller, measure) and the scope of the controller's technology implementation (no specific technology suggested).
YANG Data Model for Network Tester Management: The draft implementation was synchronized with the latest modifications and tested. A sandbox device was made available. Discussion focused on the adoption and which vendors are involved.
Benchmarking Network Performance in Container Infrastructure: A call was made for suggestions or additional changes. The draft is considered stable and ready for last call, pending more reviews.
Benchmarking Methodology for Segment Routing: The adoption of the draft was acknowledged. Discussions included addressing the compressed SID case and the relevance of security aspects from the SRv6 security draft in SPRING. The group discussed whether security benchmarking is in scope for BMWG. There was suggestion to engage segment routing experts from other WGs for review.
Characterization and Benchmarking for Power in Networking Devices: Updates were presented and relate it to work in the Green-BOF. The discussion focused on expanding the scope to include infrastructure-level energy efficiency metrics.
SRv6 Service Benchmarking Guideline: A new network-level benchmarking approach was presented, extending from single-device benchmarking. Different service types and more complex topologies were discussed.
Benchmarking Methodology for Source Address Validation: A new draft proposing a standard methodology for evaluating intradomain and interdomain source address validation was presented.
Benchmarking IoT Devices: Anna Maria Mandalay presented work on benchmarking privacy and security for IoT devices, including a problem statement regarding the lack of standards for regulations such as the Cyber Resilience Act in Europe.

Decisions and Action Items

MLR Search Draft: The chairs will discuss whether to proceed with last call after the IETF meeting, following up with the authors. Action: Chairs to decide on last call.
Container Infrastructure Draft: Authors to contact NFV-related RFC authors for review. Action: Authors to contact relevant RFC authors.
Segment Routing Draft: Authors to address compressed SID cases. Cross-post the draft on SPRING and SRV6 Ops mailing lists to engage segment routing experts. Action: Authors to address open issues and cross-post.
Power in Networking Devices Draft: Consider expanding the scope to include infrastructure-level energy efficiency metrics. Authors to incorporate new dynamic metrics discussed by green-BOF. Action: Authors to evaluate expanding scope and add metrics.
New Drafts: Libin and Anna are encouraged to continue working on their drafts. Action: All interested to contact Anna if keen to work on the draft for IoT.

Next Steps

Authors of drafts to address open issues and incorporate feedback from the meeting.
Working group to review drafts and provide comments on the mailing lists.
Chairs to push for more reviews of drafts considered ready for last call.
Authors to connect with experts in other working groups (e.g., SPRING, SRV6 Ops, IoT Ops) to solicit feedback.
Update the milestones according to the discussion today and after the meeting relating to the working group last call of some documents.