Administrators Guide
ParaStation5 Administrators Guide
Info@par-tec.com
Table of Contents
Problem different groups of nodes are seen as up or down
Introduction
History of ParaStation
About this document
Libraries
Technical overview
Runtime daemon
Kernel modules
License
Installation
Prerequisites
Hardware
Directory structure
Software
Kernel version
Man
Installation via RPM packages
Getting the ParaStation5 RPM packages
Mpi2, mpi2-intel, mpi2-pgi, mpi2-psc
Installing the RPMs
Compiling the ParaStation5 packages from source
File Version
Installing the documentation
Etc/init.d/xinetd reload
ParaStation entries
# rpm -Uv psdoc-5.0.0-1.noarch.rpm
Installing MPI
Further steps
# rpm -Uv psmpi2.5.0.0-1.i586.rpm
Uninstalling ParaStation5
ParaStation5 Administrators Guide
Copy template
Configuration
Configuration of the ParaStation system
Define Number of nodes
Enable optimized network drivers
# /opt/parastation/bin/testconfig
Hostname id HWType runJob starter accounter
Testing the installation
# /opt/parastation/bin/psiadmin -s -c list
# /opt/parastation/bin/testnodes -np nodes
Insight ParaStation5
ParaStation5 pscom communication library
# echo 10 /proc/sys/ps4/state/ResendTimeout
Directory /proc/sys/ps4/state
# cat /proc/sys/ps4/state/connections
Controlling process placement
Directory /proc/sys/ps4/local
Using the ParaStation5 queuing facility
Using non-ParaStationapplications
Exporting environment variables for a task
Controlling ParaStation5 communication paths
Export LDPRELOAD=/opt/parastation/lib64/libp4tcp.so
PSPP4S or PSPP4SOCK
Authentication within ParaStation5
Pspshm or Pspsharedmem
Export PSPLIB=/opt/parastation/lib64/libpscomopenib.so
Parallel shell tool
Homogeneous user ID space
Single system view
Nodes and CPUs
Tok2env
Integration with AFS
Integrating external queuing systems
PSIRARGPRE0=/some/path/env2tok
Multicasts
# UseMCast
Using ParaStation accounting
Copying files in parallel
Route add -net 224.0.0.0 netmask 240.0.0.0 dev ethX
Changing the default ports for psid8
Using ParaStation process pinning
Using memory binding
Spawning processes belonging to all groups
Port
Troubleshooting
Problem psiadmin returns error
Problem node shown as down
Problem different groups of nodes are seen as up or down
Problem cannot start parallel task
Problem bad performance
Problem cannot start process on frontend
Problem pssh fails
Problem psid does not startup, reports port in use
Problem processes cannot access files on remote nodes
Reference Pages
ParaStation5 Administrators Guide
Description
InstallDir inst-dir , InstallationDir inst-dir
Parastation.conf
Parameters
Stopscript
Setupscript
Startscript
Statusscript
Mvapi
P4sock
Openib
Elan
Accounter
NrOfNodes num
Node node17 16 HWType ethernet p4sock starter yes runJobs no
$GENERATE 1-96 node$0,2 $0
LogLevel num
SelectTime time
DeadInterval num
MCastGroup group-num
DataSize size
Core size
CPUTime time
MemLock size
Proc
Processes maxprocs
CPUmap map
RdpClosedTimeout ms
RdpTimeout ms
StatusTimeout ms
RdpResendTimeout ms
Errors
See also
ParaStation5 Administrators Guide
Psiadmin
Synopsis
Options
Standard Output
Standard Error
Standard Input
Extended description
Exit
All
Allproc cnt count
Hardware
Count hw hw
Down
Load
Rdp
Summary max max
Quit
Group nodes
Accounters nodes
User nodes
Maxproc nodes
HandleOldBins nodes
Master nodes
FreeOnSuspend nodes
NodesSort nodes
Cpumap nodes
Rlrss nodes
RdpClosedTimeout nodes
RdpTimeout nodes
StatusTimeout nodes
RdpResendTimeout nodes
Resolve nodes
Restart nodes
Selecttime time nodes
Psiddebug mask nodes
Pattern Name Description
HandleOldBins 0 1 nodes
Rdpmaxretrans val nodes
RdpClosedTimeout ms nodes
RdpTimeout ms nodes
StatusTimeout ms nodes
RdpResendTimeout ms nodes
Normal
Files
Quiet
Verbose
Psid
Configfile=file
Debug=level
Logfile=file
Num
Testconfig
Filename
? , --usage Show a help message
ParaStation5 Administrators Guide
Cnt num
Testnodes
Np num
Map
ParaStation5 Administrators Guide
Testpse
Testpse -npnum
ParaStation5 Administrators Guide
Net
P4stat
Sock
?,--help
ParaStation5 Administrators Guide
P4tcp
Add
Delete
ParaStation5 Administrators Guide
Psaccounter
Pattern Description
?, --help
Dumpcore
Coredir=dir
Var/account/yyyymmdd Accounting files, one per day
Psaccview
Lg,--ltotgroup
Lj,--ljobs
Lu,--ltotuser
Ls,--ltotsum
Cputime
Cpuweight
Aqtime
End
Initialization file
Mlisten
ParaStation5 Administrators Guide
Appendix A. Quick Installation Guide
# /opt/parastation/bin/psiadmin psiadmin add
# chkconfig -a /etc/init.d/parastation
Testing
Appendix B. ParaStation license
Page
Page
Page
Appendix C. Upgrading ParaStation4 to ParaStation5
Building and installing ParaStation5 packages
# psiadmin -s
Changes to the runtime environment
Page
Glossary
ARP
See ParaStation Logger
To share a common address space within a node
ParaStation5 Administrators Guide