| Login | |
Some frequently used PBS commands:
command description
qstat list information about queues and jobs
qstat -q list all queues on system
qstat -Q list queue limits for all queues
qstat -a list all jobs on system
qstat -au userid list all jobs owned by user userid
qstat -s list all jobs with status comments
qstat -r list all running jobs
qstat -f jobid list all information known about specified job
qstat -Qf queue list all information known about specified queue
qstat -B List summary information about the PBS server
qsub submit a batch job to PBS
qsub jobscript submit jobscript to PBS
qsub -I submit an interactive-batch job
qsub -l limit=value submit a job with a resource limit request, where
limit is the resource being limited, and
value is the specified limit of the resource
qsub -q queue submit job directly to specified queue
xpbs Graphical User Interface to PBS
fqstat List information about all jobs (running and queued)
and also total resources being utilized presently.
Note: On Redwood certain commands have been supplemented to support the second instance of PBS. Redwood and PBS
In March 2005 Redwood underwent a major upgrade, 128 more 1.3 GHz Itanium2 processors
were added to the existing 64 900 Mhz Itanium2 processors. So now effectively Redwood consists of
224 processors.
PBS resources on the qsub and qalter commands
cput Maximum amount of CPU time used by all processes in the job.
Units: time.
file The largest size of any single file that may be created by the job.
Units: size
mem Maximum amount of virtual memory used by all concurrent processes in
the job. Units: size.
ncpus The number of processors requested (only for shared memory mulitprocessors).
Units: unitary.
nice The nice value under which the job is to be run.
Units: unitary.
nodemask A bit mask specifying the nodes (a pair of processors) to be
associated with this job. This resource is intended for use by PBS to
optimize processor allocation and direct use of this field by the job
owner is discouraged. Units: bit mask.
pcput Maximum amount of CPU time used by any single process in the job.
Units: time.
pmem Maximum amount of virtual memory used by any single process in the job.
Units: size.
walltime Maximum amount of real time during which the job can be in the running
state. Units: time.
workingset Maximum amount of physical memory used by any single process of the
job. Units: size.
Examples:
On the qsub command line:
qsub -l cput=300,mem=24MW myscript
Within the script file:
#PBS -l cput=300
#PBS -l mem=24MW
ck_jobs_usage for Redwood and Sweetgum
This script helps determine the resource usage per job on Redwood and Sweetgum only.
[fmathew@redwood ~]$ ck_jobs_usage
CPU Req'd Use'd Use'd
JobID Name User Queue CPU Percent Memory Memory Vmem
----- ---------- -------- -------- --- ------- --------------- --------------- ---------------
309 qs3 r0697 Red-2 2 405 1.0000gb 0.2555gb 0.8559gb
310 qs2pa r0697 Red-2 2 198 1.0000gb 0.3262gb 0.8559gb
388 och3_no2 r0720 Red-2 2 373 1.0000gb 0.0893gb 1.0025gb
437 no2_cl r0720 Red-2 2 198 1.0000gb 0.4966gb 1.0025gb
438 no2_ch3 r0720 Red-2 2 199 1.0000gb 0.4595gb 1.0025gb
521 ts21b_c r0720 Red-2 2 199 1.0000gb 0.3764gb 1.0025gb
299 octmcna2wt r0759 Red-2 2 156 1.0000gb 0.2178gb 0.5918gb
399 sb3lcctzm0 r0811 Red-2 2 737 1.0000gb 0.0945gb 1.4165gb
322 si15d1B r0822 Red-2 2 194 1.0000gb 0.0626gb 1.1856gb
323 si15d1Al r0822 Red-2 2 199 1.0000gb 0.8141gb 1.1856gb
480 si12ihim r0822 Red-2 2 176 1.0000gb 0.8142gb 1.1856gb
389 M2_7071727 r0870 Red-2 2 589 1.0000gb 0.1029gb 1.4165gb
508 triad_ts4 r0871 Red-2 2 615 1.0000gb 0.1222gb 1.4405gb
36 achesrr r0898 Red-2 1 100 1.0000gb 0.0990gb 0.9043gb
370 actco_sr_r r0898 Red-2 1 100 1.0000gb 0.0642gb 0.9043gb
514 frag1_srs2 r0898 Red-2 1 100 1.0000gb 0.0648gb 0.9041gb
515 frag3_srs2 r0898 Red-2 1 104 1.0000gb 0.0671gb 0.9041gb
308 qa2pa r0929 Red-2 2 200 1.0000gb 0.1226gb 0.8559gb
329 si15d1k r0940 Red-2 2 198 1.0000gb 0.8142gb 1.1855gb
Where CPU is the number PBS shows as assigned for that job. Reminder that on Redwood a user may ask for fewer cpus than PBS assigns due to cpuset allocations.
The CPU Percent is the “average percentage usage of on CPU.” For example if CPU Percent is 50 then a
job has used 50% of one CPU --- if the CPU Percent is 300 then the job has used an average of 3 CPUs.
Note: After The upgrade on Redwood, ck_jobs_usage is supplemented by ck_jobs_usage2 for the second PBS instance, see here for more details. BR> Example of PBS Commands executed on Sweetgum
sweetgum 1% qstat
Job id Name User Time Use S Queue
---------------- ---------------- ---------------- -------- - -----
93.sweetgum g98test3 chwang 0 Q small
sweetgum 2% qstat -q
server: sweetgum
Queue Memory CPU Time Walltime Node Run Que Lm State
---------------- ------ -------- -------- ---- --- --- -- -----
large-8p -- -- -- -- 0 0 2 E R
res-gate -- -- -- -- 0 0 -- E R
SM-defR -- 288:00:0 -- -- 0 0 12 E R
MM-defR -- 288:00:0 -- -- 0 0 6 E R
LM-defR -- 288:00:0 -- -- 0 0 4 E R
LM-XR -- 672:00:0 -- -- 0 0 1 E R
LM-8P -- 672:00:0 -- -- 0 0 1 E R
LM-16P -- 672:00:0 -- -- 0 0 1 E R
--- ---
0 0
sweetgum 2% qstat -Q
Queue Max Tot Ena Str Que Run Hld Wat Trn Ext Type
---------------- --- --- --- --- --- --- --- --- --- --- ----------
large-8p 2 0 yes yes 0 0 0 0 0 0 Execution
res-gate 0 0 yes yes 0 0 0 0 0 0 Route
SM-defR 12 0 yes yes 0 0 0 0 0 0 Execution
MM-defR 6 0 yes yes 0 0 0 0 0 0 Execution
LM-defR 4 0 yes yes 0 0 0 0 0 0 Execution
LM-XR 1 0 yes yes 0 0 0 0 0 0 Execution
LM-8P 1 0 yes yes 0 0 0 0 0 0 Execution
LM-16P 1 0 yes yes 0 0 0 0 0 0 Execution
sweetgum 5% qstat -a
sweetgum.mcsr.olemiss.edu:
Req'd Req'd Elap
Job ID Username Queue Jobname SessID NDS TSK Memory Time S Time
--------------- -------- -------- ---------- ------ --- --- ------ ----- - -----
93.sweetgum.mcs chwang LM-XR g98test3 -- -- 4 10mb 00:20 Q --
sweetgum 6% qstat -f 93.sweetgum.mcsr.olemiss.edu
Job Id: 93.sweetgum.mcsr.olemiss.edu
Job_Name = g98test3
Job_Owner = chwang@sweetgum.mcsr.olemiss.edu
job_state = Q
queue = LM-XR
server = sweetgum.mcsr.olemiss.edu
Checkpoint = u
ctime = Thu Aug 10 14:11:47 2000
Error_Path = sweetgum.mcsr.olemiss.edu:/ptmp/chwang/g98test3.e93
Hold_Types = n
Join_Path = oe
Keep_Files = n
Mail_Points = be
mtime = Thu Aug 10 14:11:47 2000
Output_Path = sweetgum.mcsr.olemiss.edu:/ptmp/chwang/g98test3.log
Priority = 0
qtime = Thu Aug 10 14:11:47 2000
Rerunable = True
Resource_List.cput = 00:20:00
Resource_List.mem = 10mb
Resource_List.ncpus = 4
Resource_List.pmem = 64mb
Variable_List = PBS_O_HOME=/users/mcsr/chwang,PBS_O_LOGNAME=chwang,
PBS_O_PATH=/usr/bin:/usr/local/appl:/usr/local/appl/mpich/bin:/usr/loc
al/appl/HDF4.1r2/bin:/usr/local/appl/gamess:.:/usr/sbin:/usr/bsd:/sbin:
/bin:/usr/bin/X11:/usr/local/bin:/usr/local/appl/bin:/usr/local/appl/ms
c/bin:/usr/local/appl/matlab/bin:/usr/freeware/bin,
PBS_O_MAIL=/var/mail/chwang,PBS_O_SHELL=/bin/csh,PBS_O_TZ=CST6CDT,
PBS_O_HOST=sweetgum.mcsr.olemiss.edu,PBS_O_WORKDIR=/ptmp/chwang,
PBS_O_QUEUE=res-gate
etime = Thu Aug 10 14:11:47 2000
sweetgum 7% qstat -B
Server Max Tot Que Run Hld Wat Trn Ext Status
---------------- --- --- --- --- --- --- --- --- ----------
sweetgum.mcsr.ol 0 4 4 0 0 0 0 0 Active
| |
