…MySQL 5.6, when the workload fits entirely in memory, because there is no extra cost for memory access at random locations versus…MySQL 5.5 MySQL 5.6 MySQL 5.6 w/ read_rnd_bufer_size=4M MariaDB 5.5 MariaDB 5.5 w/ mrr_buffer_size=4M Created_tmp_disk_tables 1 1 1 1 1…
Post: Death match! EBS versus SSD price, performance, and QoS
….4 43% 1 72040.6 sda3 25.5 2.7 1.6 3.8 0.2 1.5 47% 1 78041.7 sda3 25.5 1.2 0.7 4.5 0.1 0.4 46% 0 84042.8 sda3 24.4 0.7 0.5 4….1 4.5 0.1 0.0 4.6 4% 0 12.1 sdj1 0.5 0.1 4.4 0.0 0.0 1.8 3% 0 13.1 sdj1 0.7 5.1…
Post: Improved InnoDB fast index creation
…also inefficient. Percona Server as of versions 5.1.56 and 5.5.11 allows utilizing fast index creation ….00 sec) mysql> ALTER TABLE t MODIFY v VARCHAR(4); Query OK, 4194304 rows affected (1 min 11.12… creation as reported by SHOW TABLE STATUS versus 265 MB index size with the optimization disabled…
Post: Distributed set processing performance analysis with ICE 3.5.2pl1 at 20 nodes.
… I compared the performance of 20 m1.large machines cold, versus hot. I did not record the cold results on the… too. I tested a 20 node m1.large system at 1,2,4,8,16 and 32 threads of concurrency. The following… concurrency test: $ cat start_bench #!/bin/bash if [ "$1" = "" ] then workers=1 else workers=$1 fi if [ "$2" = "" ] then iterations=3 else iterations…
Comment: Database problems in MySQL/PHP Applications
…will be saved and used. > 4. Not using relations This …versus the value added by having the benefit available. Making a blanket statement that using relations is bad is unwise. > 5. The n+1 pattern This probably should rather be called Not using > Join. This is typical error. On other hand in MySQL…
Post: Where does HandlerSocket really save you time?
… months back had some really good profile results for libmysql versus libhsclient (starting at slide 15). Somebody in the audience at… parsing) HandlerSocket (5.1.56) samples%symbol name 189462.9918btr_search_guess_on_hash 158532.5034vfprintf 87451.3810row_search_for_mysql 70211.1087buf…
Comment: New MySQL Community Release Policies published
… I read your columns 3 or 4 times so far, and all …microsecond resolution for slow queries is in 5.1 already and should come out soon. …wrote on the speed with which MySQL accepts patches. But, believe it or … disagree with your comparison of community versus enterprise. Same holds true for comparison…
Comment: MySQL 5.5.8 - in search of stability
… is enabled the write pattern is: 1) find pages to write 2) …write 4) use background IO threads to do random writes in place 5) fsync the random writes Note that step #5 is done even when O_DIRECT is used. See http://bugs.mysql…mysqlha.blogspot.com/2009/06/buffered-versus-direct-io-for-innodb.html …

