Skip to content
Snippets Groups Projects
Commit 90a2ab79 authored by Morris Jette's avatar Morris Jette
Browse files

Mostly cosmetic changes to FAQ web page

parent 8ddadea5
No related branches found
No related tags found
No related merge requests found
...@@ -178,13 +178,18 @@ launch a shell on a node in the job's allocation?</a></li> ...@@ -178,13 +178,18 @@ launch a shell on a node in the job's allocation?</a></li>
Free Open Source Software (FOSS) does not mean that it is without cost. Free Open Source Software (FOSS) does not mean that it is without cost.
It does mean that the you have access to the code so that you are free to It does mean that the you have access to the code so that you are free to
use it, study it, and/or enhance it. use it, study it, and/or enhance it.
These reasons contribute to Slurm (and FOSS in general) being subject to
active research and development worldwide, displacing proprietary software
in many environments.
If the software is large and complex, like Slurm or the Linux kernel, If the software is large and complex, like Slurm or the Linux kernel,
then its use is not without cost. then while there is no license fee, its use is not without cost.</p>
If your work is important, you'll want the leading Slurm experts at your <p>If your work is important, you'll want the leading Slurm experts at your
disposal to keep your systems operating at peak efficiency. disposal to keep your systems operating at peak efficiency.
While Slurm has a global development community incorporating leading edge While Slurm has a global development community incorporating leading edge
technology, <a href="http://www.schedmd.com">SchedMD</a> personnel have developed technology, <a href="http://www.schedmd.com">SchedMD</a> personnel have developed
most of the code and can provide competitively priced commercial support. most of the code and can provide competitively priced commercial support.
SchedMD works with various organizations to provide a range of support
options ranging from remote level-3 support to 24x7 on-site personnel.
Customers switching from commercial workload mangers to Slurm typically Customers switching from commercial workload mangers to Slurm typically
report higher scalability, better performance and lower costs.</p> report higher scalability, better performance and lower costs.</p>
...@@ -629,13 +634,13 @@ or <b>--distribution</b>' is 'arbitrary'. This means you can tell slurm to ...@@ -629,13 +634,13 @@ or <b>--distribution</b>' is 'arbitrary'. This means you can tell slurm to
layout your tasks in any fashion you want. For instance if I had an layout your tasks in any fashion you want. For instance if I had an
allocation of 2 nodes and wanted to run 4 tasks on the first node and allocation of 2 nodes and wanted to run 4 tasks on the first node and
1 task on the second and my nodes allocated from SLURM_NODELIST 1 task on the second and my nodes allocated from SLURM_NODELIST
where tux[0-1] my srun line would look like this.<p> where tux[0-1] my srun line would look like this:<br><br>
<i>srun -n5 -m arbitrary -w tux[0,0,0,0,1] hostname</i><p> <i>srun -n5 -m arbitrary -w tux[0,0,0,0,1] hostname</i><br><br>
If I wanted something similar but wanted the third task to be on tux 1 If I wanted something similar but wanted the third task to be on tux 1
I could run this...<p> I could run this:<br><br>
<i>srun -n5 -m arbitrary -w tux[0,0,1,0,0] hostname</i><p> <i>srun -n5 -m arbitrary -w tux[0,0,1,0,0] hostname</i><br><br>
Here is a simple perl script named arbitrary.pl that can be ran to easily lay Here is a simple perl script named arbitrary.pl that can be ran to easily lay
out tasks on nodes as they are in SLURM_NODELIST<p> out tasks on nodes as they are in SLURM_NODELIST.</p>
<pre> <pre>
#!/usr/bin/perl #!/usr/bin/perl
my @tasks = split(',', $ARGV[0]); my @tasks = split(',', $ARGV[0]);
...@@ -663,9 +668,9 @@ foreach my $task (@tasks) { ...@@ -663,9 +668,9 @@ foreach my $task (@tasks) {
print $layout; print $layout;
</pre> </pre>
We can now use this script in our srun line in this fashion.<p> <p>We can now use this script in our srun line in this fashion.<br><br>
<i>srun -m arbitrary -n5 -w `arbitrary.pl 4,1` -l hostname</i><p> <i>srun -m arbitrary -n5 -w `arbitrary.pl 4,1` -l hostname</i><br><br>
This will layout 4 tasks on the first node in the allocation and 1 <p>This will layout 4 tasks on the first node in the allocation and 1
task on the second node.</p> task on the second node.</p>
<p><a name="hold"><b>21. How can I temporarily prevent a job from running <p><a name="hold"><b>21. How can I temporarily prevent a job from running
...@@ -926,11 +931,10 @@ $ srun -p mic ./hello.mic ...@@ -926,11 +931,10 @@ $ srun -p mic ./hello.mic
<br> <br>
<p> <p>
Slurm supports requeue jobs in done or failed state. Use the Slurm supports requeue jobs in done or failed state. Use the
command: command:</p>
<p align=left><b>scontrol requeue job_id</b></p> <p align=left><b>scontrol requeue job_id</b></p>
</head> </head>
</p> <p>The job will be requeued back in PENDING state and scheduled again.
The job will be requeued back in PENDING state and scheduled again.
See man(1) scontrol. See man(1) scontrol.
</p> </p>
<p>Consider a simple job like this:</p> <p>Consider a simple job like this:</p>
...@@ -957,12 +961,10 @@ $->squeue ...@@ -957,12 +961,10 @@ $->squeue
JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON) JOBID PARTITION NAME USER ST TIME NODES NODELIST(REASON)
10 mira zoppo david R 0:03 1 alanz1 10 mira zoppo david R 0:03 1 alanz1
</pre> </pre>
<p> <p>Slurm supports requeuing jobs in hold state with the command:</p>
Slurm supports requeuing jobs in hold state with the command:
<p align=left><b>'scontrol requeuehold job_id'</b></p> <p align=left><b>'scontrol requeuehold job_id'</b></p>
The job can be in state RUNNING, SUSPENDED, COMPLETED or FAILED <p>The job can be in state RUNNING, SUSPENDED, COMPLETED or FAILED
before being requeued. before being requeued.</p>
</p>
<pre> <pre>
$->scontrol requeuehold 10 $->scontrol requeuehold 10
$->squeue $->squeue
...@@ -1929,6 +1931,6 @@ sacctmgr delete user name=adam cluster=tux account=chemistry ...@@ -1929,6 +1931,6 @@ sacctmgr delete user name=adam cluster=tux account=chemistry
<p class="footer"><a href="#top">top</a></p> <p class="footer"><a href="#top">top</a></p>
<p style="text-align:center;">Last modified 25 April 2014</p> <p style="text-align:center;">Last modified 30 April 2014</p>
<!--#include virtual="footer.txt"--> <!--#include virtual="footer.txt"-->
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment