|
These statistics are specific to the performance of the InnoDB storage engine. This engine type, while more heavy-weight than the default MyISAM storage engine, is useful when transactions are needed and for tables which are frequently updated. Since InnoDB does row-level, instead of table-level, locking, often times multiple INSERTs or UPDATEs can run at the same time as long as they affect different data (ie rows). As a result, while queries generally take longer as a whole when using InnoDB, more of them can generaly take place without needing to lock the tables near as often. InnoDB Stats
Currently, the only information available here is the amount of RAM (both in megabytes and as a percentage) that is being used by the InnoDB Buffer Pool. More information will be added in the future. The Buffer Pool is basically a large buffer that stores both data and indexes. As a result, for a system that uses many InnoDB tables, it is best to devote a liberal amount of RAM to the cause. Of course, if the amount of RAM used by the pool is low, you could likely decrease the amount of RAM allocated to it and use the RAM elsewhere. If, on the other hand, the RAM used is very high, and close to the total amount of RAM devoted to the pool, you may want to consider increasing the amount of RAM devoted to the cause. You can change the amount of RAM allocated to the pool by adjusting the innodb_buffer_pool_size variables in MySQL. |