Today I like to talk about the _PX_use_large_pool parameter. This parameter perform that queries slaves to use the large pool for their work, it is more a way of “divide and conquer” the global memory, when automates memory models aren’t used. The main benefice of this is isolate the shared pool for other process distinct of queries slave.
You can monitor from which area we allocate memory for PX msg buffer by query the v$sgastat view, like as show in the next example;
select * from v$sgastat where name ='PX msg pool';
Please, remember this is a hidden parameter and you must very carefull about use it in production enviroment.
Your comments and suggestions are welcomed.
HTH – Antonio NAVARRO