ZanderXu commented on PR #5430: URL: https://github.com/apache/hadoop/pull/5430#issuecomment-1705055179
> @ZanderXu Thanks for your works. Great catch here! Do you take benchmark if any possible performance decrease here when parallel fsimage loading? Almost no performance lost, because this sync is just used for new users. ``` [32 threads get names 1000000 times from 10000 different names], 311(ms) with sync, 268(ms) without sync. 43(ms) [32 threads get names 1000000 times from 10000 different names], 270(ms) with sync, 269(ms) without sync. 1(ms) [32 threads get names 1000000 times from 10000 different names], 299(ms) with sync, 268(ms) without sync. 31(ms) [32 threads get names 1000000 times from 10000 different names], 291(ms) with sync, 268(ms) without sync. 23(ms) [32 threads get names 1000000 times from 10000 different names], 319(ms) with sync, 270(ms) without sync. 49(ms) ``` -- 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]
