Originally posted by: RaviW

Need help identifying identical metric in  HMC rest Processed Metrics for pool cpu used for a managed system.

At the moment we calculate it by;

lslparutil -r pool -m <frame> 

For each time interval determine the following:
i.    Pool Cycles utilized (during the interval)
ii.    Time Cycles (during the interval)
Processors utilized during the interval = (i) / (ii).

Please help identifying the same metric - Processors utilized for a managed system physical processor pool from HMC rest process metrics.

Originally posted by: dyutiman

Hi,

Thanks for reaching out.

As part of PCM we collect Processor metrics only for Shared Processor Pool and we do not collect it for Physical Processor Pool.

- thanks

Originally posted by: RaviW

Thanks for the reply.

I found processor metrics for ManagedSystem PCM under systemFirmwareUtil and serverUtil. Can you please explain what these are?

"sampleType": "ManagedSystem",
        "systemFirmwareUtil": {
          "utilizedProcUnits": [
            0
          ],
          "assignedMem": [
            3840
          ]
        },
        "serverUtil": {
          "processor": {
            "totalProcUnits": [
              6
            ],
            "utilizedProcUnits": [
              0.072
            ],
            "availableProcUnits": [
              4.7
            ],
            "configurableProcUnits": [
              6
            ]
          }

Originally posted by: dyutiman

Hi,

The 'utilizedProcUnits' under 'systemFirmwareUtil' denotes the proc units utilized by Phyp firmware since the time system is in operating/standby state..

Under 'serverUtil'

'totalProcUnits' is the total number of proc units installed on the system.
'utilizedProcUnits' is the total proc units utilized by all lpars for the particular time interval.
'availableProcUnits' is the number of proc units available for assignment.
'configurableProcUnits' is the number of Proc units which are configurable

totalProcUnits minus configurableProcUnits provides the unlicensed proc units plus the no of processors guarded, processors inactive due to failure.

Please let me know if these help.

 - thanks

Originally posted by: RaviW

Can you please explain  'configurableProcUnits' - 'availableProcUnits' provides?

Originally posted by: dyutiman

Hi,

configurableProcUnits is the total number of proc units that can be assigned to partitions or shared processor pool's reserve capacity, where as availableProcUnits is the number of free proc units available for usage (after some proc units are assigned to partitions and shared processor pools' reserved capacity).

So the difference between configurableProcUnits & availableProcUnits will give the number of proc units already assigned to partitions and shared processor pools' capacity .

Please let me know if this helps.

 - thanks

Originally posted by: RaviW

Thanks for the reply.

'utilizedProcUnits' is the total proc units utilized by all lpars for the particular time interval. - does this contains VIOs processor utilized or is it just lpars processor utilized? 

We are trying to calculate a metric from PCM which is close to lslparutil -r pool (Physical Processor utilized).

Originally posted by: dyutiman

Hi,

Yes, all lpars include VIOs also, because underneath a Vios is also a lpar.

 - thanks