zhztheplayer commented on code in PR #9446: URL: https://github.com/apache/incubator-gluten/pull/9446#discussion_r2068186846
########## gluten-flink/docs/Flink.md: ########## @@ -105,9 +108,35 @@ bin/sql-client.sh -f data-generator.sql TODO ## Performance -Using the data-generator example, it shows that for native execution, it can generate 10,0000 -records in about 60ms, while Flink generator 10,000 records in about 600ms. It runs 10 times faster. -More perf cases to be added. +We are working on supporting the [Nexmark](https://github.com/nexmark/nexmark) benchmark for Flink. +Now the q0 has been supported. + +Results show than running with gluten can be 2.x times faster than Flink. Review Comment: 👍 It's a bit counter-intuitive. We can achieve 2x speed up even without batch processing? IIUC the current code is still row-based? -- 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]
