I would like to get more clarity on the Data Grid Agent execution. In my use case, I expect lots of parallel Map Grid Agents execution across different partition to achieve parallel logic execution closer to the data.
But there are few things which are not very clear to me. It would be a great help if WXS experts can clarify or point to right documentation section.
Here some of the questions I have....
1. How many threads per MapGridAgent instance per partition will be executed concurrently ?
2. How to tune the thread allocation (apart from minThreads & maxThreads which is at the Container level, not partition level) ?
3. Let us say we have X containers and Y partition on each containers, when a client submits N Keys for a MapGridAgent instance, how many true Threads will be spanned from client side and how many concurrent threads will be executed on grids? (Assuming keys are equally distributed across partition).
Note : We understand that the true performance is tied to the actual CPU cores available.
NOTICE: developerWorks Community will be offline May 29-30, 2015 while we upgrade to the latest version of IBM Connections. For more information, read our upgrade FAQ.
This topic has been locked.