MySQL Tunning

Take this documentation as is, not as a definitive guide to tunning MySQL. There're a lot of information and knowledge should be read before change MySQL configuration. The documentation has been created based on experice shared by OpenKM server administrator users.

Useful links should read:

• MySQL: InnoDB Performance Tuning Tips
• MySQL: The MySQL Query Cache
• InnoDB - High disk write I/O on ibdata1 file and ib_logfile0

User case 1

Server information

• OpenKM Community Version
• Virtualized Windows Server 2008
• CPU: Intel Xeon X3470 @2,93GHz
• RAM: 7GB
• JVM settings for Tomcat: -Xms 3072 -Xmx 3072 -XX:PermSize=256m -XX:MaxPermSize=512m
• Repository is 1GB

MySQL my.ini changes

• innodb_buffer_pool_size=1792M
• innodb_additional_mem_pool_size=16M
• innodb_log_file_size=256M
• innodb_log_buffer_size=8M

Using MySQL server InnoDB engine with the default configuration parameters in my.ini file can cause a serious performance bottleneck in applications with high database read write activity. MySQL should have enough RAM to be able to cache the repository and avoid the high Disk I/O activity.

The most important parameters are:

• innodb_buffer_pool_size
• innodb_additional_mem_pool_size
• innodb_log_file_size
• innodb_log_buffer_size

According to database experts, innodb_buffer_pool_size should occupy the 70 percent of RAM of a dedicated database server!!!

Before changing these values, please read MySQL documentation carefully. Before making any changes you should make sure that mysql server has been stopped correctly (check server log), otherwise the innodb will become corrupted and mysql server won't restart

Considerations Repository is 1GB so 1792M is enough RAM for innodb_buffer_pool_size. The larger your repository is the more memory you need to allocate to innodb_buffer_pool_size. If you are using x86 mysql server you cannot allocate RAM larger than 2GB.

Information collected from OpenKM forum post.

User case 2

Server information

• OpenKM Community Version
• CPU: Intel Xeon E5606 @2.13GHz
• RAM: 4GB
• System: Windows server 2008, 64bit
• JAVA_OPTS=-Xms256m -Xmx1024m -XX:PermSize=256m -XX:MaxPermSize=512m -Djava.awt.headless=true -Dfile.encoding=utf-8
• Database size:2.5G
• Repository size:40GB
• 100,000 nodes(folders + documents)

MySQL my.ini changes

• query_cache_size=0
• table_open_cache=2000
• tmp_table_size=16M
• thread_cache_size=9
• myisam_max_sort_file_size=100G
• myisam_sort_buffer_size=32M
• key_buffer_size=8M
• read_buffer_size=64K
• read_rnd_buffer_size=256K
• sort_buffer_size=256K
• innodb_additional_mem_pool_size=16M
• innodb_log_buffer_size=8M
• innodb_buffer_pool_size=1042M

Results

Before optimization the login time was about 45-55 seconds with a lot of hard-disk I/O usage. After the mysql optimization login takes about 10 seconds.

Observation: with more RAM will be possible get a more higger mysql buffer and probably better perfomance

Information collected OpenKM Forum post