PAR Technologies V5 manual ParaStation5 Administrators Guide, Info@par-tec.com

Page 2

ParaStation5 Administrator's Guide

ParaStation5 Administrator's Guide

Release 5.0.5

Copyright © 2002-2010 ParTec Cluster Competence Center GmbH

April 2010

Printed 7 April 2010, 14:11

Reproduction in any manner whatsoever without the written permission of ParTec Cluster Competence Center GmbH is strictly forbidden.

All rights reserved. ParTec and ParaStation are registered trademarks of ParTec Cluster Competence Center GmbH. The ParTec logo and the ParaStation logo are trademarks of ParTec Cluster Competence Center GmbH. Linux is a registered trademark of Linus Torvalds. All other marks and names mentioned herein may be trademarks or registered trademarks of their respective owners.

ParTec Cluster Competence Center GmbH

Possartstr. 20

D-81679 München

Phone +49-89-99809-0

Fax +49-89-99809-555

http://www.par-tec.com

<info@par-tec.com>

Please note that you will always find the most up-to-date version of this technical documentation on our Web site at http://www.par-tec.com/support.php.

Share your knowledge with others. It's a way to achieve immortality.

— Dalai Lama

Image 2

Contents Administrators Guide ParaStation5 Administrators Guide Info@par-tec.comTable of Contents Problem different groups of nodes are seen as up or down Introduction History of ParaStationAbout this document Libraries Technical overviewRuntime daemon Kernel modulesLicense Hardware InstallationPrerequisites Kernel version Directory structureSoftware Man Installation via RPM packagesGetting the ParaStation5 RPM packages Mpi2, mpi2-intel, mpi2-pgi, mpi2-pscFile Version Installing the RPMsCompiling the ParaStation5 packages from source ParaStation entries Installing the documentationEtc/init.d/xinetd reload # rpm -Uv psdoc-5.0.0-1.noarch.rpm Installing MPIFurther steps # rpm -Uv psmpi2.5.0.0-1.i586.rpmUninstalling ParaStation5 ParaStation5 Administrators Guide Copy template ConfigurationConfiguration of the ParaStation system Define Number of nodesHostname id HWType runJob starter accounter Enable optimized network drivers# /opt/parastation/bin/testconfig Testing the installation # /opt/parastation/bin/psiadmin -s -c list # /opt/parastation/bin/testnodes -np nodesInsight ParaStation5 ParaStation5 pscom communication library# cat /proc/sys/ps4/state/connections # echo 10 /proc/sys/ps4/state/ResendTimeoutDirectory /proc/sys/ps4/state Controlling process placement Directory /proc/sys/ps4/localExporting environment variables for a task Using the ParaStation5 queuing facilityUsing non-ParaStationapplications Controlling ParaStation5 communication paths Export LDPRELOAD=/opt/parastation/lib64/libp4tcp.soPSPP4S or PSPP4SOCK Authentication within ParaStation5Pspshm or Pspsharedmem Export PSPLIB=/opt/parastation/lib64/libpscomopenib.soParallel shell tool Homogeneous user ID spaceSingle system view Nodes and CPUsTok2env Integration with AFSIntegrating external queuing systems PSIRARGPRE0=/some/path/env2tokMulticasts # UseMCast Using ParaStation accountingCopying files in parallel Route add -net 224.0.0.0 netmask 240.0.0.0 dev ethXChanging the default ports for psid8 Using ParaStation process pinningUsing memory binding Spawning processes belonging to all groupsPort Problem node shown as down TroubleshootingProblem psiadmin returns error Problem different groups of nodes are seen as up or down Problem cannot start parallel taskProblem bad performance Problem cannot start process on frontendProblem pssh fails Problem psid does not startup, reports port in useProblem processes cannot access files on remote nodes Reference Pages ParaStation5 Administrators Guide Description InstallDir inst-dir , InstallationDir inst-dirParastation.conf ParametersStopscript SetupscriptStartscript StatusscriptMvapi P4sockOpenib ElanAccounter NrOfNodes numNode node17 16 HWType ethernet p4sock starter yes runJobs no $GENERATE 1-96 node$0,2 $0LogLevel num SelectTime timeDeadInterval num MCastGroup group-numDataSize size Core sizeCPUTime time MemLock sizeProc Processes maxprocs CPUmap mapRdpClosedTimeout ms RdpTimeout msStatusTimeout ms RdpResendTimeout msErrors See alsoParaStation5 Administrators Guide Options PsiadminSynopsis Standard Output Standard ErrorStandard Input Extended descriptionAllproc cnt count ExitAll Hardware Count hw hwDown LoadQuit RdpSummary max max Group nodes Accounters nodesUser nodes Maxproc nodesHandleOldBins nodes Master nodesFreeOnSuspend nodes NodesSort nodesCpumap nodes Rlrss nodesRdpClosedTimeout nodes RdpTimeout nodesStatusTimeout nodes RdpResendTimeout nodesResolve nodes Restart nodesSelecttime time nodes Psiddebug mask nodesPattern Name Description HandleOldBins 0 1 nodes Rdpmaxretrans val nodes RdpClosedTimeout ms nodes RdpTimeout ms nodesStatusTimeout ms nodes RdpResendTimeout ms nodesNormal FilesQuiet VerbosePsid Logfile=file Configfile=fileDebug=level Num TestconfigFilename ? , --usage Show a help messageParaStation5 Administrators Guide Cnt num TestnodesNp num MapParaStation5 Administrators Guide Testpse Testpse -npnumParaStation5 Administrators Guide Net P4statSock ?,--helpParaStation5 Administrators Guide Delete P4tcpAdd ParaStation5 Administrators Guide Psaccounter Pattern Description?, --help DumpcoreCoredir=dir Var/account/yyyymmdd Accounting files, one per dayPsaccview Lg,--ltotgroup Lj,--ljobsLu,--ltotuser Ls,--ltotsumCputime CpuweightAqtime EndInitialization file Mlisten ParaStation5 Administrators Guide Appendix A. Quick Installation Guide Testing # /opt/parastation/bin/psiadmin psiadmin add# chkconfig -a /etc/init.d/parastation Appendix B. ParaStation license Page Page Page Appendix C. Upgrading ParaStation4 to ParaStation5 Building and installing ParaStation5 packages# psiadmin -s Changes to the runtime environmentPage Glossary ARPSee ParaStation Logger To share a common address space within a node ParaStation5 Administrators Guide