Networking InfiniBand Cluster Bring-up Procedure

Bring-up Process Checklist

Stage

Entry Criteria

Done Criteria

Cluster Planning

-

A well defined planned topology and confirmed with actual cluster topology

Network Deployment

Non updated, default firmware/software images

All network components are upgraded and aligned to the desired versions.

Configuration

Switches are with the proper firmware/software versions

  • Switch configuration completed

  • UFM configuration completed

  • Other optional configurations completed

Cluster Verification

All previous sections completed successfully

  • SM logs are verified and all issues addressed

  • UFM Fabric Health report is cleared of errors/alarms

  • The specified UFM Telemetry indicators comply the evaluation criteria as specified here

  • BER issues monitored

  • UFM events and alarms are monitored and addressed

  • All links are up and with valid quality as described in UFM Telemetry​

Performance Testing

Cluster verification stage successfully completed

  • HPC-X package successfully installed and ClusterKit is working

  • ClusterKit run results are as expected according to 'Results Verification' section

  • Optional - results were visualized and analyzed as described in 'Visualize results' section

Last updated: