Suboptimal performance due to high utilization of the system level components

In some cases, the CPU or memory utilization on an IBM Storage Scale node is higher than 90%. Such heavy utilization can adversely impact the system performance as it affects the cycles allocated to the IBM Storage Scale daemon service.

Problem identification

On the node, issue an Operating System command such as top or dstat to verify whether the system level resource utilization is higher than 90%. The following example shows the sample output for the dstat command:

# dstat 1 10


----total-cpu-usage---- -dsk/total- -net/total- ---paging-- ---system--
usr sys idl wai hiq siq| read  writ| recv  send|  in   out | int   csw
  0   0 100   0   0   0|7308k 9236k|   0     0 |   0     0 | 812  3691
  0   0 100   0   0   0|   0     0 |3977B 1038B|   0     0 | 183   317
  1   2  98   0   0   0|   0     0 |2541B  446B|   0     0 | 809   586
  0   1  99   0   0   0|   0     0 |4252B  346B|   0     0 | 427   405
  0   0 100   0   0   0|   0     0 |3880B  346B|   0     0 | 196   349
  0   0 100   0   0   0|   0     0 |3594B  446B|   0     0 | 173   320
  1   1  98   0   0   0|   0     0 |3969B  446B|   0     0 | 692   662
  0   0 100   0   0   0|   0   116k|3120B  346B|   0     0 | 189   312
  0   0 100   0   0   0|   0     0 |3050B  346B|   0     0 | 209   342
  0   0 100   0   0   0|   0  4096B|4555B  346B|   0     0 | 256   376
  0   0 100   0   0   0|   0     0 |3232B  346B|   0     0 | 187   340

Problem resolution and verification

If the system level resource utilization is high, determine the process or application that contributes to the performance issue and take appropriate action to minimize the utilization to an acceptable level.