AstroWiki

User Tools

Site Tools

Trace:
computing_drop

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
computing_drop [2009/04/24 10:06] – leeuwencomputing_drop [2009/05/06 17:37] (current) – leeuwen
Line 1: Line 1:
-==== DROP cluster ====+====== DROP cluster ======
  
   *Drop is:   *Drop is:
-     *    With 7 nodes at 8 cores each, DROP is designed for non-interactive queue-based parallel supercomputing.  +     *    drop.astron.nl 
- +     *    With 6 nodes at 8 cores each, DROP is designed for non-interactive queue-based parallel supercomputing.  
-*    Drop is not: +  *Drop is not: 
-     *    For anything that is interactive, uses less +     *    For anything that is interactive, uses less than 8 parallel cores per job, is not queue-able, or generally takes less than a few nights to run on your desktop, you had better use one of the stand-alone servers. 
-than 8 parallel cores per job, is not queue-able, or generally takes +
-less than a few nights to run on your desktop, you had better use one +
-of the stand-alone servers. +
      *    Drop has only scratch storage. Move important data to a different server. Anything on /data/ can be deleted without notice.      *    Drop has only scratch storage. Move important data to a different server. Anything on /data/ can be deleted without notice.
      *    Kindly note that this cluster comes 'as is' with no support from IT, Joeri or Jason.      *    Kindly note that this cluster comes 'as is' with no support from IT, Joeri or Jason.
 +     *    The login node 'drop' is also the file server. Do not run jobs there, in for- or background.
 +
 +  *Come see Joeri for an account
 +
 +  *File systems
 +    */export/home/<user>
 +      *RAID6, 20GB/user quota, nothing ever auto-deleted
 +    */data (crossmounted over all nodes) = RAID0, unlimited space
 +      *RAID0, 5TB, no quotas
 +      *heaviest users will have older data auto-deleted by semi-random script, so better clear it out yourself
 +
 +  * Login/master/fileserver node is "drop", nodes are called c0 to c5
 +  * Queueing implemented to balance loads. Only use queue, do not start processing on your own.
 +  * Cluster runs [[http://www.centos.org/|CentOS5]] (a version of RHEL) under [[http://www.rocksclusters.org/roll-documentation/base/5.1/|Rocks 5.1]]
  
-* Come see Joeri for an account 
-* File systems 
-  * /export/home/<user> 
-    * RAID6, 20GB/user quota 
-  * /data (crossmounted over all nodes) = RAID0, unlimited space 
-    * RAID0, 5TB, no quotas 
-    * heaviest users will have older data auto-deleted by semi-random script, so better clear it out yourself 
computing_drop.1240567600.txt.gz · Last modified: 2009/04/24 10:06 by leeuwen