jackylee-ch commented on issue #11050: URL: https://github.com/apache/gluten/issues/11050#issuecomment-5087822029
> [@Yohahaha](https://github.com/Yohahaha) just noted this. Yes, we adjusted openCostInBytes, but it can't work as expected. So we have to introduce a straightforward way to solve the issue. We encountered similar issues when initially rolling out Gazelle internally, particularly in scenarios involving a large number of small files that created a "long-tail" effect. We currently address this using the following strategies: 1. Ensuring sufficient concurrency: Automatic configuration of `openCostInBytes`, calculated dynamically based on concurrency levels and total file size; 2. Limiting the number of small files per partition: Restricting the maximum number of files allowed in a single partition via configuration (defaulting to 10). -- 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]
