EIDF GPU Service

The EIDF GPU Service provides a containerised processing platform accelerated by NVIDIA GPUs to support scalable data processing and AI workloads.

Availability 

This service is available for University of Edinburgh researchers and DDI Programme partners.  

Please use the EIDF Application Portal to request access. You can find help and guidance on the process under "Service Access" in the EIDF user documentation. 

EIDF Application Portal

EIDF documentation

Service profile 

As of 18 April 2025, the EIDF GPU Service is composed of the following GPU types:

  • 16 NVIDIA H200
  • 136 NVIDIA H100 80GB
  • 224 NVIDIA A100 GPUs in various multi-use and single-use GPU configurations. 

Node configurations are reviewed and updated during service operation. The EIDF GPU Service employs Kubernetes and storage is provided by CephFS. 

A standard project namespace has the following initial quota (subject to ongoing review): 

  • CPU: 100 Cores
  • Memory: 1TiB
  • GPU: 12 

Note these quotas are maximum use by a single project, and that, during periods of high usage, Kubernetes jobs maybe queued waiting for resource to become available on the cluster. 

EIDF GPU Service documentation

Introductory Materials for Containers 

To make best use of the service, an understanding of containerisation is recommended. An introduction tutorial to containers for computational environments is available through the Carpentries workshop.

Carpentries workshop