DokuWiki

User Tools

Site Tools

Trace:
cluster:86

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
cluster:86 [2010/04/29 18:54]
hmeij 		cluster:86 [2010/05/13 18:29] (current)
hmeij
Line 9: Line 9:
  
 First we need to assess our usage of our clusters and if the cloud can support that. We also need to stress that we are a small liberal arts college, primarily undergraduate, with less than 3,000 students and perhaps 450 active faculty. We do not need to do cloud computing because we're going to scale upwards to 10,000 nodes.  We also have a relatively small cluster user base (50 accounts), representing many different interests groups as opposed to one large community using a small set of tools.  First we need to assess our usage of our clusters and if the cloud can support that. We also need to stress that we are a small liberal arts college, primarily undergraduate, with less than 3,000 students and perhaps 450 active faculty. We do not need to do cloud computing because we're going to scale upwards to 10,000 nodes.  We also have a relatively small cluster user base (50 accounts), representing many different interests groups as opposed to one large community using a small set of tools. 
 +
 +As I like to tell vendors whom quote me impressive statistics like  "1,500 HPC users across 17 buildings mostly doing the same thing, all tied together with their HPC solution" ... we have 50 users in 2 buildings all doing different things. 
  
 ==== Usage #1 ==== ==== Usage #1 ====
Line 28: Line 30:
 The current problems we encounter are: The current problems we encounter are:
  
-  * Home directory disk space requirement, we need more than 5 TB (out of the $298K grant, reserve $50K for a 48TB (raw) Nexsan SataBeast?)+  * Home directory disk space requirement, we need more than 5 TB (out of the $250K grant, reserve $50K for a 48TB (raw) Nexsan SataBeast?)
   * Fast scratch space, we have none (10 TB Lustre filesystem, carve out of SataBeast)   * Fast scratch space, we have none (10 TB Lustre filesystem, carve out of SataBeast)
   * Establish a data archive for users rather than have multiple copies (10 TB, carve out of SataBeast)   * Establish a data archive for users rather than have multiple copies (10 TB, carve out of SataBeast)
   * Only 16 out of 36 nodes on Infiniband   * Only 16 out of 36 nodes on Infiniband
-  * Need more nodes with small memory footprint+  * Need more nodes with small memory footprint, or more medium (12gb) nodes so small jobs can be spread wide.
   * Need more moderate memory footprint nodes (actually we need to get gaussian/linda fixed on sharptail's AMD Opteron chip set)   * Need more moderate memory footprint nodes (actually we need to get gaussian/linda fixed on sharptail's AMD Opteron chip set)
   * We need a database server   * We need a database server
cluster/86.1272567254.txt.gz ยท Last modified: 2010/04/29 18:54 by hmeij

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki