Parallel Concurrent Processing Server Concepts:
Parallel Concurrent Processing can be setup if customer is having a RAC environment so that the load can be distributed across. PCP should have minimum 2 Concurrent Nodes (node A and node B).
As parallel concurrent processing has 2 Concurrent node setup (node A and node B), each node is distributed with several concurrent managers. We assign primary and secondary node to each concurrent manager, so that the concurrent manager is started on its primary node. If primary node fails, then the concurrent manager failover to its secondary node.
The concurrent manager migrates back to its primary node once that node is available and after starting Internal Monitor process.
Internal Concurrent Manager
The Internal Concurrent Manager will run on one node at any given point of time and it can activate or deactivate any concurrent manager on any node. Since the Internal Concurrent Manager must be active at all times, it needs high fault tolerance. To provide this fault tolerance, parallel concurrent processing uses Internal Monitor Processes.
Internal Monitor Processes
Internal Monitor is to monitor the Internal Concurrent Manager and to restart that manager should it fail. You can look at node name where the Internal Monitor Process is running and every node will have its own Internal Monitor Process, but we still have to define primary and a secondary node to all Internal Monitor Process to ensure failover protection.
Log and Output File Access
One need to define a shared location to save log and out files of both the concurrent nodes. The concurrent log and output files should be accessible from any node.
For best training, contact me at firstname.lastname@example.org