hurwitzlab/launch-yourself

Name: launch-yourself

Owner: Hurwitz Lab

Description: Template script and instructions for using the TACC Launcher on Stampede.

Created: 2017-10-04 22:06:48.0

Updated: 2017-10-04 22:16:28.0

Pushed: 2017-11-06 15:32:13.0

Homepage: null

Size: 7

Language: Shell

GitHub Committers

UserMost Recent Commit# Commits

Other Committers

UserEmailMost Recent Commit# Commits

README

launch-yourself

Template script and instructions for using the TACC Launcher on Stampede.

Installation

Install the HurwitzLab fork of the TACC Launcher in your Stampede home directory.

n2.stampede(4)$ git clone https://github.com/hurwitzlab/launcher.git

To get an idea of how to use the launcher, clone this repository to some directory on Stampede. For example:

n2.stampede(103)$ git clone https://github.com/hurwitzlab/launch-yourself.git
ialized empty Git repository in /work/04658/jklynch/launch-yourself/.git/
te: Counting objects: 27, done.
te: Compressing objects: 100% (23/23), done.
te: Total 27 (delta 11), reused 13 (delta 4), pack-reused 0
cking objects: 100% (27/27), done.
n2.stampede(105)$ cd launch-yourself/
n2.stampede(106)$ ls -l
l 2
------ 1 jklynch G-814141  240 Oct  4 17:45 joblist
------ 1 jklynch G-814141 1059 Oct  4 17:45 launcher.job
------ 1 jklynch G-814141 1068 Oct  4 17:45 LICENSE
------ 1 jklynch G-814141 4179 Oct  4 17:45 README.md

Replace <developer>@email.arizona.edu with your email address on the following line of launcher.job:

TCH --mail-user <developer>@email.arizona.edu
Submit launcher.job to a single node

Submit launcher.job to SLURM with -N 1 to get 1 node. When the job completes you will have a file named something like launch-yourself.job.o8688690 which shows what happened.

n2.stampede(107)$ sbatch -N 1 launcher.job
-------------------------------------------------------------
          Welcome to the Stampede Supercomputer
-------------------------------------------------------------

eservation for this job
Verifying valid submit host (login2)...OK
Verifying valid jobname...OK
Enforcing max jobs per user...OK
Verifying availability of your home dir (/home1/04658/jklynch)...OK
Verifying availability of your work dir (/work/04658/jklynch)...OK
Verifying valid ssh keys...OK
Verifying access to desired queue (normal)...OK
Verifying job request is within current queue limits...OK
Checking available allocation (iPlant-Collabs)...OK
itted batch job 8688703
n2.stampede(108)$ ls -l
l 24
------ 1 jklynch G-814141  240 Oct  4 17:45 joblist
------ 1 jklynch G-814141 1059 Oct  4 17:45 launcher.job
------ 1 jklynch G-814141 1717 Oct  4 17:47 launch-yourself.job.o8688703
------ 1 jklynch G-814141 1068 Oct  4 17:45 LICENSE
------ 1 jklynch G-814141 4179 Oct  4 17:45 README.md
n2.stampede(109)$ cat launch-yourself.job.o8688703
CHER_WORKDIR: /work/04658/jklynch/launch-yourself
cher: Setup complete.

--------- SUMMARY ---------------
umber of hosts:    1
orking directory:  /work/04658/jklynch/launch-yourself
rocesses per host: 4
otal processes:    4
otal jobs:         8
cheduling method:  dynamic

---------------------------------
cher: Starting parallel tasks...
cher: Task 2 running job 3 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 0 running job 1 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 1 running job 4 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 3 running job 2 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)




cher: Job 4 completed in 1 seconds.
cher: Job 1 completed in 1 seconds.
cher: Job 2 completed in 1 seconds.
cher: Job 3 completed in 1 seconds.
cher: Task 1 running job 5 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 0 running job 6 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 3 running job 7 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 2 running job 8 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)




cher: Job 5 completed in 1 seconds.
cher: Job 6 completed in 1 seconds.
cher: Job 7 completed in 1 seconds.
cher: Job 8 completed in 1 seconds.
cher: Task 1 done. Exiting.
cher: Task 0 done. Exiting.
cher: Task 3 done. Exiting.
cher: Task 2 done. Exiting.
cher: Done. Job exited without errors

cher Job Complete
Submit launcher.job to two nodes

Submit launcher.job to SLURM with -N 2 to get 2 nodes.

n2.stampede(110)$ sbatch -N 2 launcher.job
-------------------------------------------------------------
          Welcome to the Stampede Supercomputer
-------------------------------------------------------------

eservation for this job
Verifying valid submit host (login2)...OK
Verifying valid jobname...OK
Enforcing max jobs per user...OK
Verifying availability of your home dir (/home1/04658/jklynch)...OK
Verifying availability of your work dir (/work/04658/jklynch)...OK
Verifying valid ssh keys...OK
Verifying access to desired queue (normal)...OK
Verifying job request is within current queue limits...OK
Checking available allocation (iPlant-Collabs)...OK
itted batch job 8688706
n2.stampede(111)$ ls -l
l 25
------ 1 jklynch G-814141  240 Oct  4 17:45 joblist
------ 1 jklynch G-814141 1059 Oct  4 17:45 launcher.job
------ 1 jklynch G-814141 1717 Oct  4 17:47 launch-yourself.job.o8688703
------ 1 jklynch G-814141 1845 Oct  4 17:50 launch-yourself.job.o8688706
------ 1 jklynch G-814141 1068 Oct  4 17:45 LICENSE
------ 1 jklynch G-814141 4179 Oct  4 17:45 README.md
n2.stampede(112)$ cat launch-yourself.job.o8688706
CHER_WORKDIR: /work/04658/jklynch/launch-yourself
cher: Setup complete.

--------- SUMMARY ---------------
umber of hosts:    2
orking directory:  /work/04658/jklynch/launch-yourself
rocesses per host: 4
otal processes:    8
otal jobs:         8
cheduling method:  dynamic

---------------------------------
cher: Starting parallel tasks...
cher: Task 3 running job 2 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 0 running job 4 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 1 running job 1 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 2 running job 3 on c517-101.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 5 running job 7 on c517-102.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 7 running job 6 on c517-102.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 4 running job 5 on c517-102.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)
cher: Task 6 running job 8 on c517-102.stampede.tacc.utexas.edu (sleep 1 && echo $LAUNCHER_JID)




cher: Job 4 completed in 1 seconds.
cher: Job 1 completed in 1 seconds.
cher: Job 3 completed in 1 seconds.
cher: Job 2 completed in 1 seconds.
cher: Task 0 done. Exiting.
cher: Task 2 done. Exiting.
cher: Task 1 done. Exiting.
cher: Task 3 done. Exiting.




cher: Job 8 completed in 1 seconds.
cher: Job 6 completed in 1 seconds.
cher: Job 5 completed in 1 seconds.
cher: Job 7 completed in 1 seconds.
cher: Task 6 done. Exiting.
cher: Task 5 done. Exiting.
cher: Task 7 done. Exiting.
cher: Task 4 done. Exiting.
cher: Done. Job exited without errors

cher Job Complete

This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.