09818nam 22006374a 450 991080850050332120200520144314.0(CKB)1000000000522930(SSID)ssj0000277535(PQKBManifestationID)11215152(PQKBTitleCode)TC0000277535(PQKBWorkID)10234449(PQKB)11225967(Au-PeEL)EBL3306577(CaPaEBR)ebr10112593(OCoLC)560313350(CaSebORM)0738426776(MiAaPQ)EBC3306577(OCoLC)835957200(OCoLC)ocn835957200 (EXLCZ)99100000000052293020031113d2002 uy 0engurcn|||||||||txtccrBuilding a Linux HPC cluster with xCAT /[Egan Ford ... et al.]1st ed.[United States?] IBM, International Technical Support Organizationc2002xxvi, 250 p. illIBM redbooks"September 2002.""SG24-6623-00."0-7384-2677-6 Includes bibliographical references and index.Front cover -- Contents -- Figures -- Tables -- Notices -- Trademarks -- Preface -- The team that wrote this redbook -- Acknowledgements -- Become a published author -- Comments welcome -- Chapter 1. HPC clustering concepts -- 1.1 What a cluster is -- 1.1.1 High-Performance Computing cluster -- 1.1.2 Beowulf clusters -- 1.2 IBM Linux clusters -- 1.2.1 xSeries custom-order cluster -- 1.2.2 IBM eServer Cluster 1300 -- 1.2.3 The new IBM eServer Cluster 1350 -- 1.3 Making up an HPC cluster -- 1.3.1 Logical functions that a node can provide -- 1.3.2 xSeries models used in our cluster -- 1.3.3 Other cluster components -- 1.4 Software -- 1.4.1 IBM Cluster Systems Management for Linux -- Chapter 2. xCAT introduction -- 2.1 What xCAT is -- 2.1.1 Download xCAT -- 2.1.2 Directory structure -- 2.2 Installing a Linux cluster with xCAT -- 2.2.1 Planning -- 2.2.2 Hardware preparation -- 2.2.3 Management node installation -- 2.2.4 Cluster installation -- Chapter 3. Hardware preparation -- 3.1 Node hardware installation -- 3.2 Populating the rack and cabling -- 3.3 Cables in our cluster -- Chapter 4. Management node installation -- 4.1 Resources to install Red Hat Linux -- 4.2 Red Hat installation steps -- 4.3 Post-installation steps -- 4.3.1 Copy Red Hat install CD-ROMs -- 4.3.2 Install Red Hat errata -- 4.3.3 Updating third party drivers -- Chapter 5. Management node configuration -- 5.1 Install xCAT -- 5.2 Populate tables -- 5.2.1 Site definition -- 5.2.2 Hosts file -- 5.2.3 List of nodes and groups -- 5.2.4 Installation resources -- 5.2.5 Node types -- 5.2.6 Node hardware management -- 5.2.7 MPN topology -- 5.2.8 MPA configuration -- 5.2.9 Power control with APC MasterSwitch -- 5.2.10 MAC address collection using Cisco 3500-series -- 5.2.11 Console server configuration -- 5.2.12 Password table -- 5.3 Configure management node services.5.3.1 Turn off services you do not want -- 5.3.2 Configure system logging -- 5.3.3 Configure SNMP -- 5.3.4 Configure TFTP -- 5.3.5 Configure NFS -- 5.3.6 Configure NTP -- 5.3.7 Configure SSH -- 5.3.8 Configure the console server -- 5.3.9 Configure DNS -- 5.3.10 Configure DHCP -- 5.4 Final preparation -- 5.4.1 Prepare the boot files for stages 2 and 3 -- 5.4.2 Prepare the Kickstart files -- 5.4.3 Prepare the post installation directory structure -- Chapter 6. Cluster installation -- 6.1 Stage 1: Hardware setup -- 6.1.1 Network switch setup -- 6.1.2 Management Processor Adapter setup -- 6.1.3 Terminal server setup -- 6.1.4 APC MasterSwitch setup -- 6.1.5 BIOS and firmware updates -- 6.2 Stage 2: MAC address collection -- 6.3 Stage 3: Management processor setup -- 6.4 Stage 4: Node installation -- 6.4.1 Creating a template file -- 6.4.2 Creating a custom kernel RPM image -- 6.4.3 Creating a custom kernel tarball image -- 6.4.4 Installing the nodes -- 6.4.5 Post-installation -- Appendix A. xCAT commands -- Command reference -- addclusteruser - Add a cluster user -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- mpacheck - Check MPA and MPA settings -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- mpareset - Reset MPAs -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- mpascan - Scan MPA for RS485 chained nodes -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- mpasetup - Set MPA settings -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Author -- Bugs -- See also -- nodels - List node properties from tables -- Synopsis -- Description -- Options -- Author -- noderange - Generate a list of node names -- Synopsis -- Description -- Options.Environmental variables -- Files -- Example -- Bugs/features -- Author -- nodeset - Set the boot state for a noderange -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- pping - Parallel ping -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- prcp - Parallel remote copy -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- prsync - parallel rsync -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- psh - Parallel remote shell -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rcons - remote console -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- reventlog - Retrieve or clear remote hardware event logs -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rinstall - Remote network install -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rinv - Remote hardware inventory -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rpower - Remote power control -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rreset - Remote hard reset -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rvid - Remote video (VGA) -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- rvitals - Remote hardware vitals -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also.wcons - Windowed remote console -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- winstall - Windowed remote network install -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- wkill - Windowed remote console kill -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Examples -- Bugs -- Author -- See also -- wvid - Windowed remote video (VGA) -- Synopsis -- Description -- Options -- Files -- Diagnostics -- Example -- Bugs -- Author -- See also -- Appendix B. xCAT configuration tables -- site.tab -- nodelist.tab -- noderes.tab -- nodetype.tab -- nodehm.tab -- mpa.tab -- apc.tab -- apcp.tab -- mac.tab -- cisco3500.tab -- passwd.tab -- conserver.tab -- rtel.tab -- tty.tab -- Appendix C. Other hardware components -- IBM Advanced Systems Management Adapter -- Equinox ESP Terminal Servers -- iTouch Communications IR-8000 Terminal Servers -- Myrinet -- Myrinet switch layout -- Setting up the Myrinet switch -- Installing the Myrinet software -- Appendix D. Application examples -- User accounts -- MPICH -- Persistance of Vision Raytracer (POVray) -- Serial POVray -- Distributed POVray using MPI-POVray -- High Performance Linpack (HPL) -- Installing ATLAS -- Installing HPL -- Related publications -- IBM Redbooks -- Other resources -- Referenced Web sites -- How to get IBM Redbooks -- IBM Redbooks collections -- Glossary -- Index -- Back cover.This IBM Redbooks publication will guide system architects and systems engineers toward a basic understanding of cluster technology, terminology, and the installation of a Linux High-Performance Computing (HPC) cluster (a Beowulf type of cluster) into an IBM eServer Cluster 1300/Cluster 1350. This book focus on xCAT Version 1.1.0 (Extreme Cluster Administration Toolkit) for installation and administration. All nodes and components of the cluster, such as compute nodes and management nodes, are installed with xCAT. This toolkit is a collection of scripts, tables, and commands used to build and administer a Beowulf type of cluster or a farm of replicated nodes. xCAT commands and configuration files are explained in the appendixes of the book. Detailed procedures on how to properly configure the Red Hat Linux 7.3 operating system in the nodes of an HPC cluster are also presented.IBM redbooks.Parallel computersBeowulf clusters (Computer systems)Parallel computers.Beowulf clusters (Computer systems)004/.35Ford Egan1635094International Business Machines Corporation.International Technical Support Organization.MiAaPQMiAaPQMiAaPQBOOK9910808500503321Building a Linux HPC cluster with xCAT3975675UNINA