Dear Slurmites, Every so often our managers require a report on which groups have used particular compute metrics over a timeout. However as our job count increases these become increasingly difficult for sreport to generate. For example, the following simple commands are having issues completing.
sreport -P cluster AccountUtilizationByUser start=2018-10-01 end=2018-11-01 format=Accounts,Login,Used -t Hours > sreport.out sreport job SizesByAccount PrintJobCount grouping=1 cluster=spartan start=2018-10-01 end=2018-11-01 > sreportcount.out I am sure there is a trivial configuration hidden away that can manage this, however my google-fu is failing me and, as a result, I have come to the expert lists for assistance instead. Thanks in advance! -- Lev Lafayette, BA (Hons), GradCertTerAdEd (Murdoch), GradCertPM, MBA (Tech Mngmnt) (Chifley) Senior HPC Support and Training Officer +61383444193 +61432255208 Department of Infrastructure Services, University of Melbourne
