On Tuesday 22 Nov 2005 10:02, Marko Knezevic wrote: > > See below, and check that the indexes are right, or that the queries > > aren't > > negating the indexes. I can run queries against a properly indexed table > > of > > 26 million rows in less than a second on lesser hardware (ok, so the SAN > > isn't lesser). The manual also contains a lot of information on how to > > tune > > MySQL for large memory environments - the default configuration is for < > > 256 > > MB I think.# > > Thank you Duncan and Scott, we are using MySQL v.4.1.12. Table type is > MyISAM and following lines are inside my.cnf file:
$ locate my-huge.cnf Read that file - it comes in the MySQL distribution, and is for 1 - 2 GB RAM servers. One generic rule that you can try to apply is to split the data spindles from your OS spindles - so your tables reside on the high speed drives, and your OS, query logs etc all reside on another drive (and even another controller). The other rule is that spindles are good. Multiple 18 gig 10K drives will give you better performance than those 300 gig drives. One of the servers I deal with has ~270 GB of storage, but it's in 74 GB drives in RAID 1+0 in an IBM DS4300. Expensive, but more spindles and more read heads to retrieve the data. ---- Scanned by mailCritical. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
