Do you have smart search configured ?
No, we don't have smart search configured. Another clue was the appearance of "JVM GarbageCollector "java.lang.OutOfMemoryError: GC overhead limit exceeded". The case described here:
I've been able to make progress on this issue - I've modified the JAVA_OPT variable to increase the "-Xmx" parameter. From what I've read, I think that this parameter sets the maximum available for the application server. I've been able to run match jobs without the CPU utilization spiking after running jobs.
What made this a little more complicated is that we are running MDM as a service. We had to out that we need to update the "service.bat" definition to update the JAVA_OPT variable.
Are there any best practices or guidelines for configuring the number of cores/amount of RAM?
After restarting the JBOSS, were you able to deploy siperian_mrm ear file?
The issue occurs when the configuration of the EJB timer setting exceeds the limit.These are warning messages which you can ignore .For more info refer the KB#469753.
Else add the below tag in all the data-sources in standalone-full.xml and restart the server by clearing the cache.
<check-validation-connection-sql>select 1 from dual</check-validation-connection-sql>
Mark the match job as complete.
File writing operation also consumes CPU so refer the KB #469753 to avoid the warning messages. Even after rectifying that if we are still facing the issue then collect the thread dumps with the logs to analyze the issue further.