Home | Tune performance | Obtain advice from the advisors

Usage: Performance and Diagnostic Advisor


The Performance and Diagnostic Advisor runs in the JVM of appservers, using PMI data to...

  1. Check for unfortunate settings
  2. Issue recommendations

To view recommendations...

Enabling has minimal system performance impact.

Get advice

  1. Enable PMI

  2. Go to...

    ...and set...

    Number of processors Number of processors on the host.
    Calculation Interval Details within the advice messages display as averages over this interval.
    Maximum Warning Sequence Number of consecutive warnings issued before the threshold is updated. For example, if set to 10, the advisor sends ten warnings, and then sets a new threshold.
    Minimum CPU for Working System CPU level indicating a host is under production load. To tune for loads that range from 50-90% CPU utilization, set to 50.
    CPU Saturated Level at which the CPU is considered fully utilized. Thread pools and other resources are not increased, even if fully utilized.

  3. Click...

      Apply | Save | server_name

      | Performance and Diagnostic Advisor Configuration | Runtime tab | Restart

    The current warning count is reset to zero (0) for each message.

  4. Simulate a production level load with suitable numbers of concurrent users.

    JVM logs should be exception free. Advice is offered when CPU utilization exceeds a sufficiently high level.

  5. Select the check box to enable the Performance and Diagnostic Advisor.

  6. To see advisor output, go to...

    ...or look in...



  7. Update the product configuration for improved performance, based on advice.

    To get best advice, use advice generated while system is in the middle of a long test. Treat with skepticism advice profferred during ramp up and down during short tests


Advice settings
View recommendations
Enable automated heap dump generation
Obtain advice
Enable using wsadmin.sh
Start lightweight memory leak detection


Search Tips   |   Advanced Search