search

radical-collaboration / extasy-grlsd

Repository to hold the input data and scripts for the ExTASY gromacs-lsdmap work
1 stars 1 forks source link

wrong number of nodes requested on bw #62

Closed euhruska closed 6 years ago

euhruska commented 6 years ago

I requested {'project': 'bamm', 'queue': 'high', 'resource': 'ncsa.bw_aprun', 'cpu_processes': 16, 'gpus': 4, 'cores': 64, 'access_schema': 'gsissh', 'cpus': 64, ' walltime': 30}, which with 16 cpus per node are 4 nodes but on bluewaters I see that this job requested only 2 nodes. I had to use entk release 0.6.1 because https://github.com/radical-collaboration/extasy-grlsd/issues/52 isn't resolved yet. Is this issue caused by the old entk version?

vivek-bala commented 6 years ago

BW compute node has hyper-threading so consider it to have 32 cores per node.

euhruska commented 6 years ago

so should request 32 core but 16 cpus per GPU (xk/GPU nodes)?

vivek-bala commented 6 years ago

I believe so yes.

euhruska commented 6 years ago

fixed with https://github.com/radical-collaboration/extasy-grlsd/issues/52