One of our customers gave me a chance to run some benchmarks on 24-core (intel cpu based) server, and I could not miss it and ran few CPU-bound tasks there. The goal of benchmarks was investigation of InnoDB-plugin and XtraDB scalability in CPU-bound load.
MySQL Binaries Percona build10
We made new binaries for MySQL 5.0.67 build 10 which include next fixes: We addressed concerns about potential logging and statistics overhead, so now you can fully turn on / off query statistics for microslow patch and user statistics in runtime. Next variables were added:
1 2 3 4 5 6 | | slow_query_log | ON | | slow_query_log_file | /var/lib/mysql/slow_query.log | for compatibility with MySQL 5.1 and | userstat_running | OFF | to control gathering of user statistics. |
With both slow_query_log = OFF and userstat_running = OFF [...]
Using Multiple Key Caches for MyISAM Scalability
I have written before – MyISAM Does Not Scale, or it does quite well – two main things stopping you is table locks and global mutex on the KeyCache. Table Locks are not the issue for Read Only workload and write intensive workloads can be dealt with by using with many tables but Key Cache [...]
Percona build7 with latest patches
We made new binaries for MySQL 5.0.67 build 7 which include patches we recently announced. The -percona release includes:
1 2 3 4 5 6 7 8 9 10 11 12 | | innodb_check_defrag.patch | Session status to check fragmentation of the last InnoDB scan | 1.0 | Percona <info@percona.com> | GPL | The names are Innodb_scan_* | | userstatsv2.patch | SHOW USER/TABLE/INDEX statistics | V2 | Google | GPL | Added INFORMATION_SCHEMA.*_STATISTICS | | show_patches.patch | SHOW PATCHES | 1.0 | Jeremy Cole | N/A | | | innodb_io_patches.patch | Cluster of past InnoDB IO patches | 1.0 | Percona | GPL | This patch contains fixed (control_flush_and_merge_and_read, control_io-threads, adaptive_flush) | | innodb_show_hashed_memory.patch | Adds additional information of InnoDB internal hash table memories in SHOW INNODB STATUS | 1.0 | Percona <info@percona.com> | GPL | | | innodb_io_pattern.patch | Information schema table of InnoDB IO counts for each datafile pages | 1.0 | Percona <info@percona.com> | GPL | INFORMATION_SCHEMA.INNODB_IO_PATTERN | | microsec_process.patch | Adds INFOMATION_SCHEMA.PROCESSLIST with TIME_MS column | 1.0 | Percona <info@percona.com> | GPL | | | innodb_locks_held.patch | Add locks held, remove locked records in SHOW INNODB STATUS | 1.0 | Baron Schwartz <baron@xaprb.com> | GPL | Bug #29126 fix | | microslow_innodb.patch | Extended statistics in slow.log | 1.1 | Percona <info@percona.com> | GPL | | | mysqld_safe_syslog.patch | Patch allows redirect output of error.log to syslog-ng | 1.0 | Percona <info@percona.com> | GPL | Ported from Debian | | innodb_fsync_source.patch | Information of fsync callers in InnoDB | 1.0 | Google | GPL | | | innodb_show_bp.patch | show innodb buffer pool content | 1.0 | Percona <info@percona.com> | GPL | | |
and -percona-highperf release additionaly includes
1 2 | | split_buf_pool_mutex_fixed_optimistic_safe.patch | InnoDB patch to fix buffer pool scalability | 1.0 | Yasufumi Kinoshita | BSD | | | innodb_rw_lock.patch | Fix of InnoDB rw_locks | 1.0 | Yasufumi Kinoshita | BSD | |
You can download RPMs for RedHat / CentOS 4.x and 5.x for x86_64, binaries, sources and patches there
Adaptive checkpointing
Do you know that there are two limits about dirty (modified but not flushed to disk) blocks of InnoDB buffer pool? One is the limit of “amount”. The other is the limit of “age”. – limit of “amount” – As you know, buffer pool of InnoDB works as write-back cache of its datafiles. If the [...]
New SpecJAppServer results at MySQL and Sun.
As you likely have seen Sun has posted the new SpecJAppServer Results More information from Tom Daly can be found here These results are quite interesting for me as I worked on some of the previous SpecJAppServer Benchmarks several years ago while being employed by MySQL. These are great results, plus they can be relevant [...]
Drilling down to the source of the problem
I had an interesting tuning case few days ago. The system serving high traffic using Innodb tables would be stalling every so often causing even very simple queries both reads and writes taking long time to complete, with progress almost paused (dropping from thousands to tens of queries per second). On the surface the problem [...]
Living with backups
Everyone does backups. Usually it’s some nightly batch job that just dumps all MySQL tables into a text file or ordinarily copies the binary files from the data directory to a safe location. Obviously both ways involve much more complex operations than it would seem by my last sentence, but it is not important right [...]
Improved InnoDB rw_lock patch
There is patch from Google to improve SMP performance , but for some workloads it showed for us reverse scalability. E.g. update_key benchmark from sysbench. There are also results with Yasufumi’s rw_locks (http://bugs.mysql.com/bug.php?id=26442)
The performance effects of new patches
We are going to show the effects of the new patches applied to Percona HighPerf release. As you see from the following graphs, there is significant difference to normal version when the data bigger than buffer pool (right graph shows CPU usage)

