On single-partition database systems, the intra_parallel parameter must be active before this parameter can be used.
This parameter specifies the number of 4 KB buffers that are used for internal communications (messages) both among and within the database servers in a partitioned database environment.
If you have multiple logical nodes on a processor, you may find it necessary to increase the value of this parameter. You may also find it necessary to increase the value of this parameter if you run out of message buffers because of the number of users on the system, the number of database partition servers on the system, or the complexity of the applications.
If you are using multiple logical nodes, on non-AIX systems, one pool of fcm_num_buffers buffers is shared by all the multiple logical nodes on the same machine, while on AIX:
Recommendation for existing Parallel Edition customers on AIX: If you are using multiple logical nodes, the value of fcm_num_buffers you used in Parallel Edition Version 1.2 may now result in significantly more storage being used per machine. For example, a four-node multiple logical node configuration may end up with four times as many FCM buffers as before.
Re-examine the value you are using; consider how many FCM buffers in total will be allocated on the machine (or machines) where the multiple logical nodes reside. You may want to change fcm_num_buffers to account for the behavior described above.