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
doku:queues [2016/04/07 08:31] – dieterdoku:queues [2016/09/06 14:10] (current) – [Node types] ir
Line 1: Line 1:
-==== Queues ====+====== Queues ======
  
-=== Standard Queue (all.q) ===+===== Standard Queue (all.q) =====
  
 The majority of jobs use the standard queue '''all.q''' with a maximum run time of 3 days (72 hours). The majority of jobs use the standard queue '''all.q''' with a maximum run time of 3 days (72 hours).
  
-== Node types ==+==== Node types ====
  
 All nodes are configured equivalently, only a few nodes have different amounts of memory. All nodes are configured equivalently, only a few nodes have different amounts of memory.
Line 14: Line 14:
   * 1 node with 512 GB main memory and 64 cores (not in '''all.q''', but in '''highmem.q''')   * 1 node with 512 GB main memory and 64 cores (not in '''all.q''', but in '''highmem.q''')
 Jobs are scheduled by default to request nodes with at least 27 GB free memory. To override this default on the command line or in the job script you may specify: Jobs are scheduled by default to request nodes with at least 27 GB free memory. To override this default on the command line or in the job script you may specify:
-  * to allow scheduling on the 32 GB, 64 GB and 128 GB nodes +  * default: allows scheduling on 32 GB, 64 GB, or 128 GB nodes 
-    * this is the default +  * request scheduling on 64 GB or 128 GB nodes, only:
-  * to allow scheduling on the 64 GB and 128 GB nodes+
     * '''-l mem_free=50G''' (on the command line) or     * '''-l mem_free=50G''' (on the command line) or
     * '''#$ -l mem_free=50G''' (in the script),     * '''#$ -l mem_free=50G''' (in the script),
-  * to allow scheduling on 128 GB nodes only:+  * request scheduling on 128 GB nodes, only:
     * '''-l mem_free=100G''' (command line) or      * '''-l mem_free=100G''' (command line) or 
     * '''#$ -l mem_free=100G''' (job script).     * '''#$ -l mem_free=100G''' (job script).
-  * to allow scheduling on 256 GB nodes only: 
-    * see below: High Memory Queue 
  
 In order to avoid jobs with low memory requirements on nodes with 64 or 128 GB, priority adjustments are made in the queue. In order to avoid jobs with low memory requirements on nodes with 64 or 128 GB, priority adjustments are made in the queue.
  
-=== Long Queue ===+===== Long Queue =====
  
 A queue where jobs will be allowed to run a maximum of 7 days A queue where jobs will be allowed to run a maximum of 7 days
Line 37: Line 34:
 <code>#$ -q long.q</code> or submit your job with <code> qsub -q long.q <job_file></code> <code>#$ -q long.q</code> or submit your job with <code> qsub -q long.q <job_file></code>
  
-=== High Memory Queue ===+===== High Memory Queue =====
  
-Due to higher memory requests from some users, two nodes with 256 GB memory and 64 cores  +Due to higher memory requests from some users, one node with 512 GB memory and 64 cores  
-are available in the queue '''highmem.q'''.+is available in the queue '''highmem.q'''.
 The four processors utilized are AMD Opteron 6274 with 2.2GHz and 16 cores each. The four processors utilized are AMD Opteron 6274 with 2.2GHz and 16 cores each.
 These nodes show a sustained performance of about 400 GFlop/s, which compares to about four standard nodes of the VSC-2. These nodes show a sustained performance of about 400 GFlop/s, which compares to about four standard nodes of the VSC-2.
  • doku/queues.1460017884.txt.gz
  • Last modified: 2016/04/07 08:31
  • by dieter