GPU Cluster – Offering

The Research Computing Platform offers a cluster of GPU Based – Container as a Service (CaaS) 

  • Kubernetes Cluster with Run:AI as scheduler
  • up to 32 nodes
    • 2 x CPU AMD EPYC 7543 32c/64t
    • Memory 512GB
    • 8 x GPU Nvidia NVIDIA A100-SXM4-80GB

The cluster accesses the scratch High Performance Storage 

  • 2.5PiB full flash (~ 800 TiB used)
  • Dedicated to compute cluster (NFS only) 
  • Scratch storage (no replication)

If you wish to test the cluster kindly contact [email protected]