Difference between revisions of "Grid5000"

From HCL
Jump to: navigation, search
Line 2: Line 2:
  
 
== Tips and Tricks ==
 
== Tips and Tricks ==
* Access is provided via access nodes: '''access.SITE.grid5000.fr'''. As soon as you are on one of sites, access.SITE1.grid5000.fr, you can directly ssh any frontend node: '''frontend.SITE2'''.
+
* Access is provided via access nodes: '''access.SITE.grid5000.fr'''. As soon as you are on one of the sites, you can directly ssh frontend node of any other site: '''frontend.SITE2'''.
* Jobs are run from the frondend nodes, using [https://www.grid5000.fr/mediawiki/index.php/Cluster_experiment-OAR2 OAR]
+
* Jobs are run from the frondend nodes, using a PBS-like system [https://www.grid5000.fr/mediawiki/index.php/Cluster_experiment-OAR2 OAR]. Basic commands:
 +
** '''oarstat''' - queue status
 +
** '''oarsub''' - job submission
 +
*** interactive job
 +
<source lang="bash">
 +
$ oarsub -I
 +
</source>
 +
*** batch job
 +
<source lang="bash">
 +
$ oarsub path_to_batch_file
 +
</source>
 +
** '''oardel''' - job removal
 
* Each site has a separate NFS, therefore, to run an application on several sites at once, you need to copy it: scp, sftp, rsync.
 
* Each site has a separate NFS, therefore, to run an application on several sites at once, you need to copy it: scp, sftp, rsync.
 
* There is no access to Internet from computing nodes (external IPs should be registered on proxy), therefore, download/update your stuff at the access nodes.
 
* There is no access to Internet from computing nodes (external IPs should be registered on proxy), therefore, download/update your stuff at the access nodes.

Revision as of 09:22, 21 April 2010

https://www.grid5000.fr/mediawiki/index.php/Grid5000:Home

Tips and Tricks

  • Access is provided via access nodes: access.SITE.grid5000.fr. As soon as you are on one of the sites, you can directly ssh frontend node of any other site: frontend.SITE2.
  • Jobs are run from the frondend nodes, using a PBS-like system OAR. Basic commands:
    • oarstat - queue status
    • oarsub - job submission
      • interactive job
$ oarsub -I
      • batch job
$ oarsub path_to_batch_file
    • oardel - job removal
  • Each site has a separate NFS, therefore, to run an application on several sites at once, you need to copy it: scp, sftp, rsync.
  • There is no access to Internet from computing nodes (external IPs should be registered on proxy), therefore, download/update your stuff at the access nodes.