Subscribe

Total Processor Busy Overflows SDK 4.1

Hi Guys,

When using the SDK  with Perl I have noticed that the value obtained for  the total processor busy is often much larger than 100 goes up to 200 + sometimes for high utilized filers. For filers that are lightly utilized it does not overflow.

Has anyone else observed this ? Is there a possible solution or do we have to wait on an SDK update ?

% CPU utilization = (total_processor_busy at time t2 - total_processor_busy at time t1) /(cpu_elapsed_time at time t2 - cpu_elapsed_time at time t1)

Re: Total Processor Busy Overflows SDK 4.1

So NMSDK has 3 performance metrics related to the processor:

  • cpu_busy ( sysstat -x like  utilization for processor performance)
  • avg_processor_busy ( average utilization across all cores )
  • total_processor_busy (utilization across all cores, example if 2 processors are present and are both at 100%. total processor busy will show 200%)

Therefore it does not overflow. For general sysstat like cpu performance metric "cpu_busy") should be used and not total_processor_busy

Re: Total Processor Busy Overflows SDK 4.1

Thank you, this would help others as well.