Slurm on Google Cloud Platform

FAQ | Troubleshooting | Glossary

• Slurm on Google Cloud Platform
• Overview
  • SchedMD
• Cluster Configurations
  • Cloud
  • Hybrid
  • Multi-Cluster/Federation
• Upgrade to v5
• Help and Support

Overview

slurm-gcp is an open-source software solution that enables setting up Slurm clusters on Google Cloud Platform with ease. With it, you can create and manage Slurm cluster infrastructure in GCP, deployed in different configurations.

Google's HPC Toolkit, on github, can be used to manage and deploy Slurm clusters and other supporting infrastrucutre via HPC Blueprints.

See supported Operating Systems.

SchedMD

SchedMD provides professional services and commercial support to help you get up and running and stay running.

Issues and/or enhancement requests can be submitted to SchedMD's Bugzilla.

Also, join comunity discussions on either the Slurm User mailing list or the Google Cloud & Slurm Community Discussion Group.

Cluster Configurations

slurm-gcp can be deployed and used in different configurations and methods to meet your computing needs.

See HPC Blueprints for HPC Toolkit example cluster configurations that are production ready.

Cloud

All Slurm cluster resources will exist in the cloud.

See the Cloud Cluster Guide for details.

Hybrid

Only Slurm compute nodes will exist in the cloud. The Slurm controller and other Slurm components will remain in the onprem environment.

See the Hybrid Cluster Guide for details.

Multi-Cluster/Federation

Two or more clusters are connected, allowing for jobs to be submitted from and ran on different clusters. This can be a mix between onprem and cloud clusters.

See the Federated Cluster Guide for details.

Upgrade to v5

See the Upgrade to v5 Guide for details.

Help and Support

• See the slurm-gcp FAQ for help with slurm-gcp.
• See the Slurm FAQ for help with Slurm.

Please reach out to us here. We will be happy to support you!

Issues

Codespell: config, workflow + fixed typos

opened on 2023-03-08 23:48:04 by yarikoptic

I skimmed over the fixes (there were no ambiguous ones) -- seems all good but please double check ;)

Fix the support of custom machine type

opened on 2023-01-24 12:53:25 by ZiggerZZ None

Fix missing bandwidth_tier into io.tf Singularity example v4

opened on 2022-07-13 15:28:04 by thiagogenez

bandwidth_tier is missing on the partition variable of the singularity example v4 module. As a result, the following error comes up:

╷ │ Error: Invalid value for input variable │ │ on main.tf line 35, in module "slurm_cluster_network": │ 35: partitions = var.partitions │ │ The given value is not suitable for │ module.slurm_cluster_network.var.partitions declared at │ ../../modules/network/io.tf:70,1-22: incorrect list element type: attribute │ "bandwidth_tier" is required. ╵ ╷ │ Error: Invalid value for input variable │ │ on main.tf line 62, in module "slurm_cluster_controller": │ 62: partitions = var.partitions │ │ The given value is not suitable for │ module.slurm_cluster_controller.var.partitions declared at │ ../../modules/controller/io.tf:149,1-22: incorrect list element type: │ attribute "bandwidth_tier" is required. ╵ ╷ │ Error: Invalid value for input variable │ │ on main.tf line 113, in module "slurm_cluster_compute": │ 113: partitions = var.partitions │ │ The given value is not suitable for │ module.slurm_cluster_compute.var.partitions declared at │ ../../modules/compute/io.tf:59,1-22: incorrect list element type: attribute │ "bandwidth_tier" is required. ╵ ERRO[0037] 1 error occurred: * exit status 1