Hello: I am planning to increase the parallel_max_servers from its current value (32) to either 64 or 128. I am trying to figure out if I need to increase shared_pool or large_pool to avoid running out of memory. Right now my shared_pool is 464M the reserved_shared_pool is 23M, large_pool is 112M. I found a note 238680.1 where it is adviced to multiply the Buffers HWM by the parallel_execution_message_size to get the value of PX mgs pool. In my case th parallel_execution_message_size is 2152 and Buffers HWM is 503. Multiplying 2152 bty 503 I am getting 1M. So I am concluding that 32 parallel slave processes are using 1M and that I will need to add another 1M to accommodate 32 more Is that a correct logic? If not, how do I need to adjust it? thank you Gene Gurevich -- //www.freelists.org/webpage/oracle-l