Reference

Technical specifications and data for Charmed HPC, covering configurations, supported values, and cluster components.

System architecture

• Underlying projects and dependencies

Hardware and networking

• GPU resource scheduling in Slurm

• Public cloud high-speed interconnects

Monitoring

Dashboards, metrics, and log queries available when COS is integrated with a Charmed HPC cluster.

• Grafana dashboards

• Loki logs

• Prometheus alerts

• Prometheus metrics

Performance

Reference data for evaluating and tuning the performance of a Charmed HPC cluster, including benchmarks and hardware-specific metrics.

• Benchmark results on Microsoft Azure