Thanks,  this the result of show status like "qcache%"

+-------------------------+----------+
| Variable_name           | Value    |
+-------------------------+----------+
| Qcache_free_blocks      | 3330     |
| Qcache_free_memory      | 13372320 |
| Qcache_hits             | 11119149 |
| Qcache_inserts          | 1942009  |
| Qcache_lowmem_prunes    | 372898   |
| Qcache_not_cached       | 508528   |
| Qcache_queries_in_cache | 3160     |
| Qcache_total_blocks     | 9711     |
+-------------------------+----------+

Am I right at saying that the value (16Mb)is ok ?

Qcache_not_cached means the number of query results that didn't get into
the cache because of a space limitation ? 
If so, perhaps adding few more Mb would improve that number , isn't it?




On Wed, 2005-03-09 at 19:30, Dan Nelson wrote:
> In the last episode (Mar 09), Mauricio Pellegrini said:
> > Hi, I'm trying to tweak the execution time for querys on my mysql
> > server which is using InnoDB.
> > 
> > I know that there are a lot of things to check but one of them,
> > perhaps not the most important, is the query_cache_size.
> > 
> > I have a dual Xeon cpu with 4gb of ram and this is a Linux dedicated
> > server which runs only Apache and Mysql 4.1.5
> > 
> > I've declared Query_cache_size = 16M in the my.cnf file and I would
> > say that in the whole we have 60% of select querys and a 40% of
> > update querys.
> > 
> > Is Query_cache_size=16M too conservative considering that I have 4Gb
> > of ram ?
> 
> A better question is: what is your current utilization of that 16MB
> cache?  The query cache flushes any results dependant on a table when
> that table is modified, so if you really have 40% updates, chances are
> that very few queries are cachable at all.  Run
> 
>   show status like 'qcache%';
> 
> , and take a look at free_memory, inserts, and hits.
>  
> -- 
>       Dan Nelson
>       [EMAIL PROTECTED]


-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]

Reply via email to