How To Hold And Release Job in PBS
#qsig -s suspend command to suspend a job so that it won’t run. If you suspend a job, and then release it using the
#qsig -s resume command,
HOW TO HOLD THE PARTICULAR JOB.
For Example 54.
1)qstat -a 54 (list all the jobs id 54)
2)qhold 54 (hold the job 54)
3)qstat -a 54 (check the status)
4)qrls -h u 54 (release the job id 54)
5)qstat -a 54 (check the Status)
PBS Resource Management Useful Tips
1)qsub -l select=4:ncpus=1:mem=4gb -l place=free (Take the resource anywhere PLACE).
2)qsub -l select=1:ncpus=2:mem=50gb:host=zooland(Request job on specified NODE).
3)qsub -l select=2:ncpus=3:mem=6gb -l place=scatter (Request resource on different NODE).
4)qsub -l select=1:ncpus=1:mem=25gb -l place=pack:excl (Request a small number of CPUs but a large amount of memory, exclu-
sively)
User’s PBS Environment
User must be able to transfer files between hosts (e.g. via rcp or scp)
should add entry /usr/bin/scp in /etc/pbs.conf
qalter Change Attribute of the Particular Job.
PBS place statement
The place statement may be not be used without the select statement.
The place statement has this form:
-l place=[ arrangement ][: sharing ][: grouping]
where
arrangement is one of free | pack | scatter
sharing is one of excl | shared
grouping can have only one instance of group=resource
USER MAINTENANCE : flatuid
1)qsub -u:
a)flatuid attribute determines whether it assumes that identical
usernames mean identical users.
b)If true, it assumes that if UserS exists on
both the submission host and the server host, then UserS can run jobs on
that server.
#qstat - Different New Option
-t show the status of job array object and subjobs.will also show state of jobs.
-T Estimate Job Starting Time(When the job will start if it is Queue).
-J show state only of job arrays.
-p How much percentage job is completed.
-n Get List of nodes allocated to the job number(#qstat -n JOBID).
Future Article : PBS GUI
SIMPLE PBS TEST SCRIPT FILE
$ cat test1
#!/bin/bash
#PBS -N test
#PBS -l select=4:ncpus=16
cd /scratch/TEST_HOST
module load intel-cluster-studio-2013
mpirun -machinefile $PBS_NODEFILE -np 64 /bin/hostname
#qsig -s suspend command to suspend a job so that it won’t run. If you suspend a job, and then release it using the
#qsig -s resume command,
HOW TO HOLD THE PARTICULAR JOB.
For Example 54.
1)qstat -a 54 (list all the jobs id 54)
2)qhold 54 (hold the job 54)
3)qstat -a 54 (check the status)
4)qrls -h u 54 (release the job id 54)
5)qstat -a 54 (check the Status)
PBS Resource Management Useful Tips
1)qsub -l select=4:ncpus=1:mem=4gb -l place=free (Take the resource anywhere PLACE).
2)qsub -l select=1:ncpus=2:mem=50gb:host=zooland(Request job on specified NODE).
3)qsub -l select=2:ncpus=3:mem=6gb -l place=scatter (Request resource on different NODE).
4)qsub -l select=1:ncpus=1:mem=25gb -l place=pack:excl (Request a small number of CPUs but a large amount of memory, exclu-
sively)
User’s PBS Environment
User must be able to transfer files between hosts (e.g. via rcp or scp)
should add entry /usr/bin/scp in /etc/pbs.conf
qalter Change Attribute of the Particular Job.
we can change only cputime and walltime.
#qstat -u barry
Req'd Elap
Job ID User Queue Jobname Sess NDS TSK Mem Time S Time
-------- ------ ----- ------- ---- --- --- --- ---- - ----
51.south barry workq airfoil 930 -- 1 -- 0:16 R 0:01
54.south barry workq airfoil -- -- 1 -- 0:20 Q --
#qalter -l walltime=25:00 -N engine 54
JOb ID 54 is changing
1)walltime from 20 to 25 and
2)Name from airfoil to engine
#qstat -a 54
Req'd Elap
Job ID User Queue Jobname Sess NDS TSK Mem Time S Time
-------- ------ ----- ------- ---- --- --- --- ---- - ----
54.south barry workq engine -- -- 1 -- 0:25 Q --
#qstat -u barry
Req'd Elap
Job ID User Queue Jobname Sess NDS TSK Mem Time S Time
-------- ------ ----- ------- ---- --- --- --- ---- - ----
51.south barry workq airfoil 930 -- 1 -- 0:16 R 0:01
54.south barry workq airfoil -- -- 1 -- 0:20 Q --
#qalter -l walltime=25:00 -N engine 54
JOb ID 54 is changing
1)walltime from 20 to 25 and
2)Name from airfoil to engine
#qstat -a 54
Req'd Elap
Job ID User Queue Jobname Sess NDS TSK Mem Time S Time
-------- ------ ----- ------- ---- --- --- --- ---- - ----
54.south barry workq engine -- -- 1 -- 0:25 Q --
PBS place statement
The place statement may be not be used without the select statement.
The place statement has this form:
-l place=[ arrangement ][: sharing ][: grouping]
where
arrangement is one of free | pack | scatter
sharing is one of excl | shared
grouping can have only one instance of group=resource
USER MAINTENANCE : flatuid
1)qsub -u:
a)flatuid attribute determines whether it assumes that identical
usernames mean identical users.
b)If true, it assumes that if UserS exists on
both the submission host and the server host, then UserS can run jobs on
that server.
#qstat - Different New Option
-t show the status of job array object and subjobs.will also show state of jobs.
-T Estimate Job Starting Time(When the job will start if it is Queue).
-J show state only of job arrays.
-p How much percentage job is completed.
-n Get List of nodes allocated to the job number(#qstat -n JOBID).
Future Article : PBS GUI
SIMPLE PBS TEST SCRIPT FILE
$ cat test1
#!/bin/bash
#PBS -N test
#PBS -l select=4:ncpus=16
cd /scratch/TEST_HOST
module load intel-cluster-studio-2013
mpirun -machinefile $PBS_NODEFILE -np 64 /bin/hostname
0 comments:
Post a Comment