Summary

• 2 minutes

In this module, you learned:

• How to identify unhealthy high-performance computing (HPC) VMs by running several specific virtual machine (VM) health checks.
• How a single unhealthy VM can cause a tightly coupled HPC application to fail or run more slowly than expected.
• How to prepare to run a tightly coupled HPC parallel application optimally.
• How to troubleshoot HPC applications that run slowly.

Learn more

• HPC on Azure

• VM sizes for HPC

• Microsoft Message Passing Interface (MPI)

• Intel Memory Latency Checker

• CentOS HPC 7.7+

• HPC I/O benchmarks

• Fio project

• IOzone Filesystem Benchmark

• Vdbench