Skip to main content

Minutes IETF116: bmwg

Meeting Minutes Benchmarking Methodology (bmwg) WG
Date and time 2023-03-28 04:00
Title Minutes IETF116: bmwg
State Active
Other versions plain text
Last updated 2023-03-28


BMWG Session (IETF-116)
March 28, Tuesday Session II
0400 - 0600  UTC
Other Time Zones:
Room: G316     ops      bmwg

AGENDA   - 28 participants, total.

WG Status:

 - RFC: 9411, Benchmarking Methodology for Network Security Device Performance
 - WG Adoptions, and Proposals

WG Drafts:

MLR Search and YANG Model (brief status)
- One leaf removed in YANG Model, MLR discussed later.

 + Benchmarking Methodology for Stateful NATxy Gateways using RFC 4814
   Pseudorandom Port Numbers
   - comments exchanged on new version
   - Generating Multiple IP addresses are needed/proposed for the method now.
   This is not yet implemented. - WGLC?  Start after authors implement proposal
   and review.


 + Considerations for Benchmarking Network Performance in Containerized
 Infrastructures - Minh-Ngoc Tran
   - Sharing updates and progress, thanks for reviews
   - Pods are only deployed on Baremetal hosts
   - Revised deployment figures, New
   - Looking at results from Hackathon 115 some improvement with new OVS
   - Vratko - will review, then review can be part of the adoption call. maybe
   combined review with Maciek. Eduard And Gabor will also review. - ACTION to
   wait for review notifications, then WG Adoption.

 + Benchmarking Methodology for MPLS Segment Routing - Paolo Volpato
   Related Draft:
   - Paolo: describe the latest updates to address the comments received so far
   - Recap, Need to expand on RFC 5695,  Test SID list is longer than one SID
   in the RFC - Gabor and Bruno provided comments. - How is TST 007 used?
   wrong, meant TST009 - Louis Chen, Juniper, 3 Comments about: - no affect of
   certain label operations. - Why have a long list of SIDs? - Also, consider
   taking action on the QoS of the labels, for scalability. Paolo agreed to
   expand the text - Ask for Volunteer reviews to the list. pre-adoption.

 + Benchmarking Methodology for IPv6 Segment Routing - Eduard Vasilenko
   Related Draft:
   - Eduard: describe the latest updates to address the comments received so far
   - Feedback from Gabor and Bruno - big change to add encapsulations in
   transit, and more labels - Louis Chen, Juniper, asked about MicroSID is out
   of scope, not approved yet, doc could be very large if trying to include
   every possible test. - Seek Adoption via volunteer reviewers on the list -
   try for more participation. - ACTION on both SR drafts: seek review
   assignments pre-Adoption.

  + Problems and Requirements of Evaluation Methodology for Integrated Space
  and Terrestrial Networks
   - Presenter: Qi Zhang
   - updated draft with additional considerations, plus new draft
   - Can we define meaningful benchmarking that fits within BMWG's lab-only
   charter? - Key question: what limited aspect of the many-dimensional ISTN
   problem presented here could be pursued for benchmarking network DUT/SUT? -
   Qi Zhang: review of progress. - Regulators and space observations can
   provide constellation info. Need orbital parameters for testing/report. -
   Ground station and Inter-Sat connectivity need to be determined. Meshes can
   be formed among near Ground stations. Network links may be used based on
   delay calculations and link speeds. - Slide with OSPF? latencies - How
   measured? ping used. Telsat has the least number of satellites, so longest
   latencies <100ms  stiil better than GEO. NOTE: all these latency
   measurements are UN-LOADED - best case... - Plan to work on more concrete
   methods for benchmarking - Paolo: Latency tests with IPv4 connectivity? yes.
   Is IPv6 supported? consider finding this in the future. - Qi: Reg info
   changes frequently - find the representative parameters for repeatable

 + LAST - MLR Draft Update - Vratko
   - Current version of draft is out of date
   - Expanded classifications of trial outcomes
   - reviewed 8 example classification of load conditions after a number of
   trials -

 + AOB