Attendees:

Please add your name in here:

  1. Kelvin Edmison (Nokia)
  2. Ahmed El Sawaf (STC)
  3. Mark Shostak
  4. Pankaj Goyal (AT&T)
  5. Karine Sevilla (Orange)
  6. Tomas Fredberg (Ericsson)
  7. Ulrich Kleber (Huawei)
  8. Nick Chase (Mirantis)
  9. Tom Van Pelt (GSMA)
  10. Gergely Csatari (Nokia)
  11. Trevor Cooper (Intel)
  12. Petar Torre (Intel)
  13. Al Morton (AT&T)
  14. Toshiyasu Wakayama (KDDI)

Special Notes:

  • Given the limited time available, we will focus on identifying issues/actions/next steps, but not solving them right now
  • Weekly RM meetings are intended to
    • Identify owners for new Issues
    • Track open Issues
    • Address technical issues that cannot be resolved online (i.e. resolve stalls)

Agenda:



  • Guidelines for host hardware selection
    • Slides from Prague
    • Should these be included in the RM Appendix (to be informative but not normative)


  • RM Chapter 8 removal/discussion
    • Slides from Prague
    • Should these be included in the RM Appendix (to be informative but not normative)


  • Open questions
    • <TBD>


  • New Business
    • Can the RM meeting move 1 hr earlier?
    • This had been explored previously, but conditions have changed (i.e. preceding RC meeting using same zoom bridge)
      • Kelvin Edmison to explore this option.
        • Answer: yes, it can move.  This was determined too late to re-schedule this meeting, but re-scheduling will take effect for next week.

Actions:

  • General
    •  Walter Kozlowskito create Etherpad for Networking Fabric discussions for post-Baldy release
    • Walter Kozlowski call the first meeting for the Networking Fabric topic

Minutes:

  • Discussion on the Host hardware selection slides
    • Strong opinions on preserving operator choice of hardware platforms and configurations
    • Points raised that some operators have extremely good tools about workload placement/platform usage, etc and that such guidelines would impair them
    • Decision taken to not introduce this into the RM.
  • Discussion led from there into performance characterization/performance predictability
    • Discussion led to a point where RM did not specify performance at all, but RC tests are specified sufficiently well to raise and characterize performance differences per NFVI subsystem
    • Decision that for now, no test tools will be specified in RM, and that RC test scenarios will be used to characterize and report the performance of an NFVI.
  • RM Chapter 8 removal discussion did not occur due to time constraints.
  • No labels