[Ofmfwg] Sunfish meeting notes for April 4, 2025

[Lee, Peter]

  1.  H3 CXL work
     *   Skip
  2.  Documentation
     *   Russ will add additional information to the documentation and the readme in Sunfish Server repo based on the H3 CXL activity.
  3.  Github repository consolidation
     *   May need to rethink our approach. Russ attempted to do some changes and ran into issues, so backed out.
     *   DMTF and Lustre both leverage a main branch, with different branches for different things. This could be an approach we can also take. More investigation needs to be done. However, we will not pursue any changes until Russ can complete his changes.
  4.  Flux resources from Broker 0 to the compute brokers
*         Mapping of the closest CDI resources to the computes

        *   Still trying to get the full picture of how things work in Flux, but Sunfish has all the data on the resources. If we can get the information from Sunfish into Flux, things should work out.
        *   One thing to be aware of is that not all hardware managers may supply physical topology data as Redfish does not have a good way to describe it. The SNIA mockups we are using don't include physical topology. We may need to make changes to the mockups.
  1.  Flux Prolog/Epilog integration work, burst buffer deployment
     *   Have code for the burst buffer that we can use for the POC demo.
  2.  NVMeoF Sunfish Agent
     *   Code not ready yet but have the infrastructure and mockups ready. Just need to put them all together.
     *   Still target to get this done before the SNIA event at the end of April.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ofmfwg/attachments/20250404/14088d99/attachment-0001.htm>