IBM High Performance Computing Cluster Health Check

Introduction

This IBM Redbooks publication mainly provides information about IBM High Performance Computing (HPC) clusters. Therefore, in the rest part of this book, unless otherwise mentioned, the term cluster represents an IBM HPC cluster. A computer cluster is a group of connected computers that work together. In many respects, they act as a single system.

We describe the concepts, processes, and methodologies used to achieve and maintain a “healthy” state for an IBM HPC system, both in pre-production stage and production stage.In the context ...