=========================================
#fdio-meeting: FD.io CSIT project meeting
=========================================


Meeting started by mackonst_ at 14:02:57 UTC.  The full logs are
available at
http://ircbot.wl.linuxfoundation.org/meetings/fdio-meeting/2023/fd_io_csit_project_meeting/fdio-meeting-fd_io_csit_project_meeting.2023-06-07-14.02.log.html
.



Meeting summary
---------------

* Vratko Polak.  (vrpolak, 14:03:07)
* Tibor Frank  (tifrank, 14:03:24)
* mackonstan  (mackonst_, 14:03:38)
* VPP updates  (mackonst_, 14:04:54)
  * LINK:
    https://developer.nvidia.com/blog/nvidia-grace-cpu-superchip-architecture-in-depth/
    (vrpolak, 14:04:59)

* TSC updates  (mackonst_, 14:05:04)
  * Vratko reporting: last week discussion about equipment orders for
    2023, this week LFN dev&test forum so no TSC meeting  (mackonst_,
    14:07:14)
  * #info Vratko reporting: Casey brought up the new Nvidia CPU based on
    Neoverse V2 Cores: Armv9 with 4x128b SVE2 - is FD.io / CSIT
    interested in having it in the lab  (mackonst_, 14:09:39)
  * LINK:
    https://www.supermicro.com/en/pressreleases/supermicro-add-nvidia-grace-cpu-superchip-based-servers-industry-leading-portfolio
    (vrpolak, 14:11:46)
  * Dave Wallace  (dwallacelf, 14:14:48)
  * mackonstan - suggest we check with the Arm team (Honappa, Lijian) re
    availability of any server with Nvidia Crace CPU  (mackonst_,
    14:18:40)
  * Dave - just did the lfn dev&test session on VPP Strongswan plugins -
    two separate plugins by Nanotech and by Intel  (mackonst_, 14:20:55)

* AMD performance results  (mackonst_, 14:23:29)
  * Vratko - opened a new ticket for bad AMD multi-core scaling with cx5
    and rdma-core driver (mlx5 dpdk driver is fine). same combination on
    clx works fine.  (mackonst_, 14:26:11)

* previous rls csit-2302  (mackonst_, 14:26:50)
  * re performance RCAs: RSS patch is still open  (mackonst_, 14:27:21)

* csit-2306  (mackonst_, 14:27:32)
  * Tibor - rc1 testing finished, and retest of vpp 23.02 in current
    csit environment almost done, some dpdk test runs are still to do
    (mackonst_, 14:28:16)
  * Tibor - majority of issues fixed, Peter - link stability issues on
    SPR, still testing  (mackonst_, 14:33:53)
  * Tibor - report will be published using csit.fd.io dashboard
    (mackonst_, 14:36:16)
  * Tibor - static content already updated, Peter added csit environment
    updates, Vratko added methodology for reassembly test  (mackonst_,
    14:36:44)
  * Tibor - documentation - following sections require work: rls notes.
    (mackonst_, 14:36:46)

* Physical Infrastructure  (mackonst_, 14:37:16)
  * mackonstan - SPR results are looking good, thanks to Peter for all
    the work getting them up&running in time for the release - Kudos!!
    (mackonst_, 14:39:28)
  * mackonstan - SPR nat44ed performance looks lower than expected,
    Vratko and Maciek to follow up - most likely related to TRex ASTF
    behaviour  (mackonst_, 14:40:03)
  * Peter - progressing with mlx6 and mlx7 NIC installs in 3n-alt and
    icx testbeds  (mackonst_, 14:40:58)

* Virtual Infrastructure, CI/CD  (mackonst_, 14:44:05)
  * Vratko - jenkins sandbox down, ping Vanessa for update  (mackonst_,
    14:44:27)

* Environment, Framework  (mackonst_, 14:45:21)
  * Vratko - re VPP API tracking > In-progress messages - need to raise
    with VPP to change APIS from In-progress to Production  (mackonst_,
    14:47:24)
  * Vratko - re CLI without API - CSIT is using a number of features
    without APIs - need to a list to bring it up with VPP community
    (mackonst_, 14:49:50)

* Trending and Test Failures  (mackonst_, 14:50:06)

* Fd.io Events  (mackonst_, 14:53:19)
  * LFN Developer & Testing Forum - CSIT talk tomorrow Thursday 8th of
    June 6-7am PDT / 3-4pm CEST  (mackonst_, 14:54:03)



Meeting ended at 14:54:06 UTC.



People present (lines said)
---------------------------

* mackonst_ (32)
* collab-meetbot (9)
* vrpolak (3)
* tifrank (1)
* dwallacelf (1)
* jlinkes (0)
* mackonstan (0)



Generated by `MeetBot`_ 0.1.4