FelixYBW opened a new pull request, #11877: URL: https://github.com/apache/gluten/pull/11877
By default, when we create CPUThreadPoolExecutor, the LifoSemMPMCQueue is used, which uses LIFO to schedule requests. Which means it always schedule the last rowgroup request firstly. But in Table scan, we process rowgroup from index 0 which leads to wait of the first rowgroup. UnboundedBlockingQueue is FIFO queue, it can make sure the first rowgroup is scheduled firstly. A better solution is to use PriorityUnboundedBlockingQueue which can add a priority value when we add task to executors. But it needs to modify DBI. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
