Grid Engine cluster configuration
Grid Engine root directory ($SGE_ROOT ) | ${cfg.sge.root} |
Cell name ($SGE_CELL ) | ${cfg.cell.name} |
Cluster name ($SGE_CLUSTER_NAME ) | ${cfg.sge.cluster.name} |
Qmaster port ($SGE_QMASTER_PORT ) | ${cfg.sge.qmaster.port} |
Execd port ($SGE_EXECD_PORT ) | ${cfg.sge.execd.port} |
Group id range ($SGE_GID_RANGE ) | ${cfg.gid.range} |
Qmaster spool directory | ${cfg.qmaster.spool.dir} |
Global execd spool directory | ${cfg.execd.spool.dir} |
Spooling method | ${cfg.spooling.method} |
Spooling directory | ${cfg.db.spooling.dir} |
JMX port | ${cfg.sge.jmx.port} |
JVM library path | ${cfg.sge.jvm.lib.path} |
JMX SSL server keystore path | ${cfg.sge.jmx.ssl.keystore} |
Administrator mail | ${cfg.admin.mail} |
| Succeeded | Failed |
Qmaster host | ${add.qmaster.host} | ${add.qmaster.host.failed} |
Execution host(s) | ${cfg.exec.host.list} | ${add.exec.host.list.failed} |
Shadow host(s) | ${cfg.shadow.host} | ${add.shadow.host.failed} |
Admin host(s) | ${cfg.admin.host.list} | ${add.admin.host.list.failed} |
Submit host(s) | ${cfg.submit.host.list} | ${add.submit.host.list.failed} |
How to start with Grid Engine
-
Set the environment:
- if you are a csh/tcsh user:
source ${cfg.sge.root}/${cfg.cell.name}/common/settings.csh
- if you are a sh/bash/ksh user:
. ${cfg.sge.root}/${cfg.cell.name}/common/settings.sh
This will set or expand the following environment variables:
$SGE_ROOT
(always necessary)
$SGE_CELL
(if you are using a cell other than default
)
$SGE_CLUSTER_NAME
(always necessary)
$SGE_QMASTER_PORT
(if you haven't added the service sge_qmaster
)
$SGE_EXECD_PORT
(if you haven't added the service sge_execd
)
$PATH/$path
(to find the Grid Engine binaries)
$MANPATH
(to access the manual pages)
-
Submit one of the sample scripts contained in the ${cfg.sge.root}/examples/jobs directory.
qsub ${cfg.sge.root}/examples/jobs/simple.sh
or
qsub ${cfg.sge.root}/examples/jobs/sleeper.sh
-
Use the qstat command to monitor the job's behavior.
qstat -f
-
After the job finishes executing, check your home directory for
the redirected stdout/stderr files
script-name.e
job-id and
script-name.o
job-id.
The job-id is a consecutive unique integer number assigned to each job.
Administering Grid Engine
Grid Engine startup scripts:
Daemon | Location | Actions |
Qmaster |
${cfg.sge.root}/${cfg.cell.name}/common/sgemaster |
start | stop | restart |
Exec daemon |
${cfg.sge.root}/${cfg.cell.name}/common/sgeexecd |
start | stop | softstop | restart |
Startup messages can be found in SMF service log files.
You can get the name of the log file by calling svcs -l SERVICE_NAME
E.g.: svcs -l svc:/application/sge/qmaster:${cfg.sge.cluster.name}
After startup the daemons normally log their messages in their
spool directories as follows, but can be configured to use syslog.
Qmaster: |
${cfg.qmaster.spool.dir}/messages |
Exec daemon: |
execd_spool_dir/ hostname/messages |
Useful links