How much a database can push out has a lot dependencies. The first
probably being a very fast disk subsystem. If your disk can't pump out way
more than what you are trying to get, then it doesn't matter what software
or how many CPU's you have.
Disk is hardly being used. The database its self is fairly small (30K
rows?) and (from what I can tell) is all in memory. (in the current MSSQL
setup).
I could go on about dependencies on your data (1 image or many rows), your
indexes (very important for queries per second), your web server, etc.
Indexes we can add, which we have. Unfortunately we are dealing with a
packaged application which we have only some abilities to deal with the
code (a total switch to MySQL we could rewrite the code for, but we can not
optimize specific queries as there is a common SQL generator which actually
generates the SQL statement).
But I think you are more interested on if it can be done with MySQL. YES,
it can. I was just reading yesterday about Yahoo's switch to PHP and the
performance problems they had with remember.yahoo.com (?). That was there
first big project using PHP, MySQL, Apache. At first, MySQL got crushed by
the load. They added something like 20 more MySQL slaves which then got
things working under heavy load.
I see a link for this, and looked at it briefly. However it sounds like
your resources has more info. Would you still have the link?
Their hind site analysis was interesting. A big part of the reason MySQL
"failed" was poor database design and lack of indexes on key fields. Join
fields were not indexed!
DB design is depend on the application. Indexes we can do though, and have.
Thanks for this info. Info which is useful to build a case.
benji
---
Ben Spencer
Web Support
[EMAIL PROTECTED]
x 2288
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
