DokuWiki

User Tools

Site Tools

Trace:
cluster:192

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
Next revision Both sides next revision
cluster:192 [2020/02/26 18:34]
hmeij07 [Usage] 		cluster:192 [2020/02/27 15:46]
hmeij07 [Miscellaneous]
Line 4: Line 4:
 ===== EXX96 ===== ===== EXX96 =====
  
-A page for me on how these 12 nodes were build up after they arrived. To make them "ala n37" which as the test node in redoing our K20 nodes, see [[cluster:172|K20 Redo]]+A page for me on how these 12 nodes were build up after they arrived. To make them "ala n37" which was the test node in redoing our K20 nodes, see [[cluster:172|K20 Redo]] and [[cluster:173|K20 Redo Usage]] 
 + 
 +Page best followed bottom to top if interested in the whole process. 
 + 
 +The Usage section below is HPCC users wnatig to use queue ''exx96''.
  
-Page best followed bottom to top. 
  
 ==== Usage ==== ==== Usage ====
  
-The new queue ''exx96'' will be comprised of nodes ''n79-n90''.  Each node holds 4x RTX2080S gpus, 2x Xeon Silver 4214 2.2 Ghz cpus, 96 GB memory and a 1TB SSD. ''/localscratch'' is around 800 GD.+The new queue ''exx96'' will be comprised of nodes ''n79-n90''.  Each node holds 4x RTX2080S gpus, 2x Xeon Silver 4214 2.2 Ghz 12-core cpus, 96 GB memory and a 1TB SSD. ''/localscratch'' is around 800 GB.
  
-A new static resource is introduced for all nodes holding gpus. ''n78'' in queue ''amber128'' and ''n33-n37'' in queue ''mwgpu''.  The name of this resource is ''gpu4''.  Moving forward please use it instaed of ''gpu'' or ''gputest''.+A new static resource is introduced for all nodes holding gpus. ''n78'' in queue ''amber128'' and ''n33-n37'' in queue ''mwgpu'' and the nodes mentioned above.  The name of this resource is ''gpu4''.  Moving forward please use it instead of ''gpu'' or ''gputest''.
  
-The wrappers provided assume your cpu:gpu ration is 1:1 hence in your submit code you will have ''#BSUB -n 1'' and in your resource allocation line ''gpu4=1''.  If your ratio is something else you can set CPU_GPU_REQUEST, for example CPU_GPU_REQUEST=4:2 which expectas the lines ''#BSUB -n 4'' and ''gpu4=2'' in your submit script.+The wrappers provided assume your cpu:gpu ratio is 1:1 hence in your submit code you will have ''#BSUB -n 1'' and in your resource allocation line ''gpu4=1''.  If your ratio is something else you can set CPU_GPU_REQUEST. For example CPU_GPU_REQUEST=4:2 expects the lines ''#BSUB -n 4'' and ''gpu4=2'' in your submit script. Sample script at ''/home/hmeij/k20redo/run.rtx''
  
-The wrappers (78.mpich3.wrapper for n78, and n37.openmpi.wrapper for all others) are located in ''/usr/local/bin'' and will set up the environment and start these applications: amber, lammps, gromacs, matlab and namd.+The wrappers (n78.mpich3.wrapper for ''n78'', and n37.openmpi.wrapper for all others) are located in ''/usr/local/bin'' and will set up your environment and start either of these applications: amber, lammps, gromacs, matlab and namd from ''/usr/local''.
    
  
 <code> <code>
 +
 +# command that shows gpu reservations
 bhosts -l n79 bhosts -l n79
              gputest gpu4              gputest gpu4
  Total             0    3  Total             0    3
- Reserved        0.0  0.1+ Reserved        0.0  1.0
  
 +# old way of doing that
 lsload -l n79 lsload -l n79
  
Line 30: Line 36:
 n79                         ok   0.0   0.0   0.0   0%   0.0     0   0 2e+08  826G   10G   90G    3.0 n79                         ok   0.0   0.0   0.0   0%   0.0     0   0 2e+08  826G   10G   90G    3.0
  
-mdout.325288:|  Master Total CPU time:          982.60 seconds     0.27 hours  1:1 +</code>
-mdout.325289:|  Master Total CPU time:          611.08 seconds     0.17 hours  4:2 +
-mdout.326208:|  Master Total CPU time:          537.97 seconds     0.15 hours 36:4+
  
-#BSUB -n 4 +Peer to peer communication is possible (via PCIe rather than NVlink) with this hardware.  This will get rather messy in setting up.  Some quick off the cuff performance data reveals some impact. Generally in our environment the gains are not worth the effort.  Using Amber and ''pmemd.cuda.MPI''
-#BSUB -R "rusage[gpu4=2:mem=6288],span[hosts=1]" +
-export CPU_GPU_REQUEST=4:2+
  
-</code>+<code> 
 +                                                                              cpu:gpu 
 +mdout.325288:|  Master Total CPU time:          982.60 seconds     0.27 hours   1:1 
 +mdout.325289:|  Master Total CPU time:          611.08 seconds     0.17 hours   4:2 
 +mdout.326208:|  Master Total CPU time:          537.97 seconds     0.15 hours  36:4 
 + 
 +</code> 
 ==== Miscellaneous ==== ==== Miscellaneous ====
  
Line 78: Line 86:
  
 # for mwgpu/exx96 -pm=ENABLED -c=DEFAULT # for mwgpu/exx96 -pm=ENABLED -c=DEFAULT
-nvidia-smi --persistence-mode=1 +# note: turned this off, running with defaults 
-nvidia-smi --compute-mode=0+#nvidia-smi --persistence-mode=1 
 +#nvidia-smi --compute-mode=0
  
 # turn ECC off (memory scrubbing) # turn ECC off (memory scrubbing)
cluster/192.txt · Last modified: 2022/03/08 18:29 by hmeij07

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