apurtell commented on pull request #3244: URL: https://github.com/apache/hbase/pull/3244#issuecomment-844561641
I am redoing microbenchmarks with the latest patch and will update here soon. Removing the extra copy unlocked IO performance improvement from the compression. |Algorithm|Avg WAL append in ms - Before|Avg WAL append in ms - Now| |--|--|--| |None|0.290 (stdev 0.328)|unchanged| |Compression, but no value compression|0.372 (stddev 0.336)|unchanged| |Compression, value compression=SNAPPY|2.971 (stddev 2.029)|**0.0272 (stddev 0.204)**| Before, SNAPPY+copy had a 10x loss, now SNAPPY+no-copy has a 10x gain. I double checked these findings and it seems correct. I will re-run the benchmark when measuring for the other codec types for comparison. -- 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: [email protected]
