WASP and memory settings?

[Rikard Thorin]

Our UMP can be very slow and difficult to work with sometimes.

I've changed the parameter of the WASP java startup, from 2048MB to 4096MB. But in my UMP server the Java-task is stabilized around 2000MB.

My initial start-up is set to 1024MB.

What am I doing wrong?

The CPU utilization looks very good

 

Is there any more settings you can do to increase the performance of the UMP?

I'm running USM/UMP version 8.2, 64-bit Java, 10GB memory and enough CPU apparently.

[unnar]

I am having similar performance issues and I figure it cant only be the two of us that are experiencing this.

 

If I restart my UMP, it sometimes takes 45+ minutes for the wasp to come back online and I get these in the log:

Oct 19 11:34:53:703 INFO  [pool-1-thread-1, com.nimsoft.nimbus.probe.service.wasp.udm.StatusCheckThread] UDM not initialized yet. Trying to initialize.

Oct 19 11:34:53:703 DEBUG [pool-1-thread-1, com.nimsoft.nimbus.probe.service.wasp.udm.UDM] Sending InitializeDataUpdate to dashboard_engine

Oct 19 11:34:53:933 WARN  [pool-1-thread-1, com.nimsoft.nimbus.probe.service.wasp.udm.UDM] Unable to initialize data update from dashboard_engine.

Oct 19 11:34:53:933 WARN  [pool-1-thread-1, com.nimsoft.nimbus.probe.service.wasp.udm.UDM] (7) temporarily out of resources, Received status (7) on response (for sendRcv) for cmd = 'InitializeDataUpdate'

 

Have you been able to resolve your performance issues? BTW, I am running 8.3.

[Rikard Thorin]

Well it's a bit better now after upgrading the memory to 38GB, and the maximum memory in Wasp is now set to 30GB.

But still... the UMP is quite slow.

is there any new features for the UMP in 8.3 ?

[Rob van der Weijden]

Hi Rith,

 

We have similar issues. Our primary UMP has 20GB and 16 CPU`s, however we stil have long response times.

[David Michel]

With either UIM 8.2 or 8.3 if there are performance problems in UMP try re-configuring  dynamic groups into static groups.