You are viewing a plain text version of this content. The canonical link for it is here.

Posted to notifications@skywalking.apache.org by GitBox <gi...@apache.org> on 2019/06/29 06:04:05 UTC

[GitHub] [skywalking] lkxiaolou opened a new pull request #2930: One more time for heigher performance

lkxiaolou opened a new pull request #2930: One more time for heigher performance
URL: https://github.com/apache/skywalking/pull/2930
 
 
   - Why submit this pull request?
   - [ ] Bug fix
   - [ ] New feature provided
   - [x] Improve performance
   
   AtomicRangeInteger v1 is the original implementation.
   AtomicRangeInteger v2 is rewrote by me which impoves 3x~10x performance,.
   And today I'm wrote AtomicRangeInteger v3 which improves about 30% performance to the v2.
   
   The V3 algorithms is the same as v2, and I'm learning that AtomicRangeInteger V2 _value_ is a false-sharing with the _startValue_ and _endValue_, in the v3 implementation it was disabled the false-sharing with a 64 byte array to store the _value_ , so get the _startValue_ or _endValue_ can use the CPU's L1 cache line.
   

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services