This redbook documents the building of a Linux HPC cluster on IBM eserver xSeries hardware. Using this approach, and building a type of Beowulf cluster, we can achieve dramatic performance at a fraction of the price of previous solutions. HX5CMNMHKUUR
This redbook will also teach you the concepts and principals needed to build your own HPC cluster. The installation process is streamlined and simplified by the extensive use of scripts. Management tools are provided to easily manage a large number of compute nodes that remotely use the built-in features of Linux and the advanced management capabilities of the IBM eserver xSeries Service Processor Network.
Authors: Luis Ferreira (also known as “Luix”) is a Software Engineer at IBM Corp - International Technical Support Organization, Austin Center working in Linux and AIX projects, Gregory Kettmann (also known as “Greg”) is an IT Architect working for IBM Global Services, ITS, as a Subject Matter Expert in Linux Clusters, Jens Ihamo is IT Specialist at the IBM Techline in Greenock, Scotland, Andreas Thomasch is a Linux specialist and consultant at the EMEA Linux Center of Competence in Boeblingen, Germany, Steve Hill is a High Performance Computing specialist working in Hursley, England, Walter Bernocchi is an IT Specialist at the xSeries Pre-Sale Technical Support team in Italy, Jean Claude Daunois is a System Engineer in France, Eileen Silcocks is an IBM Technical Trainer in Greenock, Scotland, Jacob Chen is an IT Specialist of IBM Taiwan, Makoto Harada is an IT Engineer at PS Server, ATS IBM Japan.
Table of Contents:
- Chapter 1. Introduction
- Chapter 2. General cluster architecture
- Chapter 3. Components overview
- Chapter 4. Solution guide
- Chapter 5. Hardware preparation
- Chapter 6. Management node installation
- Chapter 7. Compute node installation
- Chapter 8. Installation of additional components