Meeting, Wednesday September 9, 2023

Participants:
- Sam Grayson
- Tren Hao
- Nathan Somers
- Jake Levine
What names should we settle on?
- GLUG/Infra cluster
Purpose and offerings of the other cluster and of our cluster?
- HPC cluster: batch execution for SIG{AIDA, Ecom, HPC}
  - SIGAIDA also has fleshed out requirements, but could be computationally intensive
    - NCSA compute cluster may be a better option
- Container execution cluster: general purpose, SIGPWNY, SIGAIDA, ACM Infra, ACM Projects
  - SIGPWNY has the most fleshed out requirements
  - ACM Projects Committee Saurav & Max leads
  - For our cluster: offer container engine, maybe VM space. We don't have many nodes and only 1 GPU.
What IPAM network for each of us to use?
- Could use same
What level of collaboration going forward?
- Get requirements documents from other SIGs, especially SIGAIDA and SIGPWNY
- Raw k8s access could be acceptable user interface/service offeirng
Next steps:
- GLUG/Infra cluster team:
  - What we want to do
    - Revise task list
      - Emphasize doing container engine first
  - How we're going to do it
    - Make them all have the same distro
    - Find candidate HTTP/HTTPS service, ask Engr IT to open external port for that service
    - Deploy persistent storage for k8s
    - Deploy k8s
- Jake:
  - Send us requirements docuemnts from SIGAIDA, SIGPWNY, Infra, Applications, others...
  - forward infra-compute applications over

Provide feedback