Do we need to have a JVM installed for an external Database server in order to retrieve data of a BLOB column into Pega?
Consider a scenario where a select query is executed to retrieve data from a typical BLOB column which is defined in a DB table that exists in an external schema (on a separate DB server).
It is known that JVM allocates a large chunk of memory in order to read the data stream transmitted from the BLOB column. However, in the case of above scenario, it is not clear whether the JVM utilized is of DB server or Pega Application Server?
Kindly clarify whether having JVM installed on an external DB server is mandatory to retrieve the BLOB data?
PRPC reads BLOBs from mapped database tables via JAVA User Defined Functions (PR_READ_FROM_STREAM). The UDFs and Java classes reside and are loaded and executed by the JVM built-in within the third-party enterprise Database server (e.g - Oracle, IBM DB2 etc.). So, the answer to your question on whether you need to have a JVM installed on an external DB server is mandatory would be "No".
In PROD,we currently have two independent databases. One DB server is part of Pega installation and in another external DB (Hitachi) some third party data is stored. We currently have Java running on both database servers. Now, we plan to migrate the third party data to AWS, and use RDS database which do not have Java. The question is - "Would it be still possible to access/read the data stored in BLOB without java running on the DB server?"