masaori335 commented on PR #9394: URL: https://github.com/apache/trafficserver/pull/9394#issuecomment-1437724766
> With my ts::scalable_shared_mutex, with 16 threads, it had worse performance that ts::shared_mutex. So maybe check the performance of this mutex with 16 threads? Please look at the table and graph above. The benchmark is measured with 1,2,4,8,16,32, and 64 threads. ts::brovo::shared_mutex shows better performance with fewer threads. > ts::scalable_shared_mutex uses the class DenseThreadId... IIUC, reducing collision between threads by using the hash function (mix32) is one of the key ideas of BRAVO. However, for ATS, we might be able to reduce the slots with your DenseThreadId, because we don't use dynamic threading. I'll try to use it and measure performance. -- 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]
