Some Query Cache fixups.

Following are some performance data for the query cache

(We got these by running the MySQL benchmark suite on a Linux Alpha

2x500 MHz with 2GB RAM and a 64MB query cache).

2x500 MHz with 2GB RAM and a 64MB query cache):

@itemize @bullet

@item

If want to disable the query cache code set @code{query_cache_size=0}.

If you want to disable the query cache code set @code{query_cache_size=0}.

By disabling the query cache code there is no noticeable overhead.

@item

If all of the queries you're preforming are simple (such as selecting a

If a table changes (@code{INSERT}, @code{UPDATE}, @code{DELETE},

@code{TRUNCATE}, @code{ALTER} or @code{DROP TABLE|DATABASE}),

then all cached queries that used this table (possibly through a

MRG_MyISAM table!) become invalid and are removed from the cache.

@code{MRG_MyISAM} table!) become invalid and are removed from the cache.

Currently all @code{InnoDB} tables are invalidated on @code{COMMIT},

in the future this will be changed so only tables changed in the

@item @code{Qcache_inserts}

 @tab Number of queries added to the cache.

@item @code{Qcache_hits}

 @tab Number of cache hits

 @tab Number of cache hits.

@item @code{Qcache_not_cached}

 @tab Number of non-cached queries

 (not cachable, or due to SQL_QUERY_CACHE_TYPE)

 (not cachable, or due to @code{SQL_QUERY_CACHE_TYPE}).

@item @code{Qcache_free_memory}

 @tab Amount of free memory for query cache.

@item @code{Qcache_total_blocks}

 @tab Total number of blocks in query cache.

@item @code{Qcache_free_blocks}

 @tab Number of free memory blocks in query cache.

@end multitable

Please note:

Total number of queries =

@code{Qcache_inserts} + @code{Qcache_hits} + @code{Qcache_not_cached}.

The query cache uses variable length blocks, so @code{Qcache_total_blocks}

and @code{Qcache_free_blocks} may indicate query cache memory fragmentation.

After @code{FLUSH QUERY CACHE} only a single (big) free block remains.

Note: Every query needs a minimum of two blocks (one for the query text

and one or more for the query results). Also, every table that is used

by a query needs one block, but if two or more queries use same table

only one block needs to be allocated.

@node Table types, Clients, Reference, Top

@chapter MySQL Table Types