probably biased towards AMD.
money is a powerful thing, which is why it should be taken with a grain
of salt as you stated.
i would guess the 12% decrease on 64bit xeon and 32% increase on 64bit
opteron is BS. why didn't they try itanium instead? ;)
David Griffiths wrote:
Anandtech has an interesting article
(http://www.anandtech.com/IT/showdoc.aspx?i=2447) on hardware for
Linux database servers.
Some very interesting conclusions:
1) Moving to 64-bit MySQL on a 64-bit Xeon actually decreases
performance by about 12% on average, while an Opteron running 64-bit
MySQL gets a 32% performance increase.
2) Innodb scales better (obviously)
3) A server with one CPU that has a dual-core Opteron (the X2 CPUs) is
faster than a server with two single-core CPUs.
4) SuSE SLES 9.1 outperforms Gentoo by about 12%
I would take Anandtech with a grain of salt - this isn't what they
normally do, and I can't verify their benchmarking was reasonably
accurate (surprised at the disks they used - one ATA, one SCSI).
David.
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
