xdmod

An open framework for collecting and analyzing HPC metrics.

The Job Performance (SUPReMM) module for Open XDMoD.

Ansible playbooks & scripts for the installation & maintenance of a Slurm based cluster at MGHPCC hosted by UMass Amherst RC

Federated XDMoD

Async Slack bot (Bolt for Python) for generating CPU, GPU, or queue wait‐time reports from XDMoD. Use /metrics to pick date range, metric, aggregate or timeseries, with optional user/group filters. Outputs Harvard‐branded charts or summaries. Includes Prometheus telemetry and Docker support.

Make container image running XDMOD on Rocky Linux 9, Singularity and apptainer

Script to generate Lustre usage metrics in JSON format for ingesting in to OpenXDMoD

Self-contained XDMoD deployment using Apptainer