Protecting Shared Storage from Bad I/O Patterns with Altair Mistral

Enterprise workload managers for high-performance computing (HPC) efficiently allocate jobs and orchestrate CPU, memory, license, and GPU sharing – but they're not designed to control the workloads' I/O patterns, an area that may be overlooked when tuning HPC-environment performance. It's easy for a job with bad I/O patterns to overload shared storage, which commonly happens when a user has tested a workflow on one or two nodes, then scales the workload without understanding how the I/O patterns scale.

Altair Mistral is the leading application monitoring tool for HPC and scientific computing, with the unrivaled ability to track I/O patterns across an HPC cluster. Altair Mistral monitors I/O, CPU, and memory, quickly locating rogue jobs and storage bottlenecks and keeping track of what's running on clusters day-to-day.