Skip to Content

Configure SAP Predictive services database pool


Adjust the database pool setting of the SAP Predictive services to control the load on your HANA instance

How-To Details

The HANA MDC instances available on the SAP Cloud Platform trial environment are limited to only 6GB of RAM.

Therefore it is really important to control the load else your HANA MDC will simply restart because not enough resources are available.

The SAP Predictive services are by default configured to run at most 4 jobs in parallel.

A job is created whenever you run one the SAP Predictive services in asynchronous mode.

The number of concurrent jobs is limited by the number of available database connections in the pool.

Luckily this parameter is exposed as one of the JMX (Java Management Extensions) parameters of the aac4paservices Java application.

It can be configured without changing any piece of code in the SAP Predictive services deployed application via the JMX Console.

Step 1: Access the JMX Console

Now that the aac4paservices is deployed and started, you will be able to access the JMX Console.

On the left side menu bar, you can expand Monitoring > JMX Console.

aac4paservices JMX Console
Step 2: Access the database pool size parameter

Now, expand > binding-java-comp/env/jdbc/DefaultDB-job > JobPoolMBean from the tree.

As you can see the default value is 4.

aac4paservices JMX Console
Step 3: Modify the database pool size parameter

Click on Operations, and enter 1 as the changeMaximumPoolSize parameter value.

aac4paservices JMX Console

Click on the execute icon aac4paservices JMX Console.

An new entry will appear in the Operation results.

Note: if you click back on Attributes, you will notice that the updated value is not displayed. You will need to refresh the page or switch to another parameter.

It is not required to restart the application after the parameter is updated. and if the application is restarted, the change will be maintained.

Next Steps

Back to top