May 23, 2012

Post: Evaluating IO subsystem performance for MySQL Needs

performance pattern compared to buffered IO (even followed by fsyncwith pretty uniform performance. Let us now do the test with single thread and write-through RAID…get multiple outstanding requests. Performance with 64 threads was …hardware I used for benchmarks – it is Dell PowerEdge 2950 with

Post: Innodb Double Write

RAID is used page may come on the stripe border requiring multiple IO requests. Same happens with Hardware RAID… much does double write buffer affect MySQL Performance ? Even though double write requires each page… allows Innodb to save on fsync()s – instead of calling fsync() for each page write …

Post: MySQL 4 to MySQL 5 Upgrade performance regressions

… at the same time you should not have hardware RAID or have one without battery backed up cache …would notice large amount of log writes and fsyncs per second which matches your hard drive capacity…your performance back. XA support has its overhead anyway so you may not get same performance as with

Post: Group commit and real fsync

… enabled (as it enables XA) Certain OS/Hardware configurations still fake fsync delivering great performance at the cost of being non ACID First… extra tests for fsync() correctness. Install RAID with battery backed up cache. This gives about the same effect as fake fsync() but you can…

Post: When would you use SAN with MySQL ?

…instead of simple RAID you use with directly attach storage which can magically improve performance ? There are…than one fsync() for the log because it internally implements XA to synchronize with binary …” which can experience “failure”. Note even if hardware fully redundant the Firmware may have a…

Comment: Tokyo Tyrant - The Extras Part I : Is it Durable?

…a separate “fast” spindle & do an fsync against just that tranlog. The …hardware-based tuning. As for indices, if the DBM (TC) allows for using clustered PK indices (esp. withRAID – are more likely to spread out across the IO system & reduce IO wait. The table bloats more quickly but the performance

Post: Testing the Group Commit Fix

…-backed cache on your RAID card). This scenario may…hardware used in the tests was a HP ProLiant DL380 G6 withwith the Group Commit Fix applied. The first graph shows the results when we have a slow fsyncwith system without writeback cache impacts performance significantly. And the similar with

Comment: Group commit and real fsync

…threaded system evaluation benchmark Running the test with following options: Number of threads: 1 …fsync() at the end of test, Enabled. Using synchronous I/O mode Doing random write test Threads started! Done. Operations performed…(avg/stddev): 0.1558/0.00 Hardware: 2×5130 2.3 Ghz, 6 Gb, 237 RAID 10 …