…patch from Google to improve SMP performance , but for some workloads it showed for us reverse … results with Yasufumi’s rw_locks (http://bugs.mysql.com/bug.php?id=26442) ThreadsStandard InnoDBGoogle smpfixYasufumi … on 8cores box, and as you see for 4+ threads we have worse results than for 2 threads…
Post: Great work Innodb Team
November 30, 2009 By 8 Comments
… interest which is Performance And Scalability. Innodb Plugin 1.0.4 had a lot of great performance improvements and 1.0.5/1.0.6 gets… be merged with MySQL Documentation which makes it much more usable and Innodb Plugin is shipped with MySQL 5.1 which makes it much easier to use which means wider community testing. Number of patches suggested by…
Post: Faster MySQL failover with SELECT mirroring
February 1, 2009 By 9 Comments
…over in both directions. Aside from MySQL Cluster, which is more special-purpose… Since we’re running Percona’s patched server, we can actually measure …4.09 6.29 site.user_achievements 1.36 1.20 0.35 site.clicks 1… insert buffer can also cause terrible performance. There are some subtleties about …
Post: New SpecJAppServer results at MySQL and Sun.
November 12, 2008 By Leave a Comment
…4 cores was enough. Does it mean the J2EE system got saturated before the database ? Or does it mean MySQL…mysql-slow.log long_query_time = 1 innodb_data_home_dir = /data/mysql…of performance. innodb_thread_concurrency = 0 Good! So Innodb performed…Innodb (and we have patches for this) Operating System…
Post: Can Innodb Read-Ahead reduce read performance ?
June 26, 2007 By 2 Comments
… OS file writes, 2038246 OS fsyncs 1 pending preads, 0 pending pwrites 286…. case it is likely these are 4 read-ahead requests, 64 pages each…. which hurt performance ? I guess we really should try Mark Callaghan patches to see… was adviseor in this project from MySQL AB side) – to improve how Innodb…
Post: XtraDB/InnoDB CPU bound benchmarks on 24cores server
December 29, 2008 By 6 Comments
… management: I tested MySQL-5.1.30 with InnoDB plugin, Xtradb-1.0.2-1, and XtraDB-1.0.2-2 (…patch). my.cnf is [mysqld] user=root binlog_cache_size=1M default_table_type=MYISAM ft_min_word_len=4…, read detailed investigation what is the reason of performance drop in next post. And last one workload …
Post: MySQL/Innodb scalability tests after fix
January 3, 2007 By 1 Comment
…higher concurrency. The results for MySQL 5.0.20a vs 5.0.32bk are also extremely interesting: 1) For single CPU …is too large to only map it to Innodb scalability patch. 2) Speedup for 5.0.32 for large number …peak performance for 5.0.32 is also much larger than for 5.0.20 – some 630 vs 540 tps. 4) …
Comment: Call for opinions: Do we need MySQL 5.0 with MySQL 5.4 performance
May 1, 2009 By 12 Comments
… often scared about freshness of the patches without getting into specifics. A lot of these performance optimization changes are rather local and so safe. It is like ant and elephant in comparison if you look at MySQL 5.1 or MySQL 5.4 changes. XtraDB is a more tricky beast mainly…
Comment: MySQL Users Conference - Innodb
May 9, 2007 By 6 Comments
Hello Peter, Thanks for the sum-up on upcoming InnoDB changes! Any performance improvements will be welcome 
About Google patches, you wrote “These are for MySQL 5.0 at this point” I suppose you meant 4.1, right?
Comment: Benchmarking single-row insert performance on Amazon EC2
May 16, 2012 By 17 Comments
…’s measurements (http://www.tokutek.com/2012/01/1-billion-insertions-%E2%80%93-the-wait-is…are perhaps 10 times bigger than they were 4 years ago when the iiBench project started, …net/~mdcallag/mysql-patch/mytools/annotate/head%3A/bench/ibench/iibench.py) is that it measures query performance too,…
