Show TOC

 Operating System Monitor Data: CPULocate this document in the navigation structure

Definition

The following data about CPU usage is displayed for every CPU, broken down as percentages by:

  • User
  • System
  • Times in which the CPU had no task to perform or was waiting for an input/output (idle)

Many factors could lead to an excessively high CPU utilization, and you should therefore perform a detailed analysis. If the problem was caused by too many active processes in the host system, you could, for example, transfer CPU-intensive programs to times when there is a lower system workload, or to other host systems. You could also increase the number of CPUs or upgrade the CPU(s).

Note

When determining the hourly values for the last 24 hours ( Previous Hours) and the last 30 days ( History), these values are averaged across all CPUs of a host.

Other Values Collected

  • Number of CPUs (only in  Snapshot)
  • Interrupts per second
  • System calls per second
  • Context switches per second
  • Average number of waiting processes for the last minute, last five minutes and the last 15 minutes

    This is the number of processes for each CPU that are in a wait queue before they are assigned to a free CPU. As long as the average remains at one process for each available CPU, the CPU resources are sufficient. As of an average of around three processes for each available CPU, there is a bottleneck at the CPU resources.

    • In connection with a high CPU usage, a high value here can indicate that too many processes are active on the server.
    • In connection with a low CPU usage, a high value here can indicate that the main memory is too small. The processes are then waiting due to excessive paging.
      Note

      When determining the hourly values for the last 24 hours ( Previous Hours) and the last 30 days ( History), the system only displays the average number of waiting processes for the last five minutes.

Additional values are also displayed, depending on which platform you are using:

  • For hosts on which IBM System p or IBM IBM i are running, a large number of values in connection with the logical partitions (LPARs) and the associated virtualization are displayed. More information: SAP Note 1131691.
  • You can also display platform-specific values for IBM z/OS; however, to do so you need to activate a corresponding data collection. This is described in SAP Note 1236078. In this case, an additional navigation tree is displayed for the monitored sysplexes (more information: Selecting Operating System Data of Systems, Instances, and Hosts).
Use

Check the following performance factors in particular:

Display Procedure

Is a CPU user constantly active?

Check whether the process is in an endless loop.

Is the average load > 3 (more than three processes are waiting for the CPU)?

Check whether all processes with high CPU usage (memlog, r3trans, nwengine, brbackup...) are necessary.

Is the usage of the CPU zero percent?

Check the analysis for the previous hours.

Note

More information about the data collected about CPU utilization: Operating System Monitor.

 Operating System Monitor start page