MySQL vs Oracle

[dozey dude]

I have been working with Oracle now for almost 10 years and I
obviously think it's probably the best and quickest database system
out there. But then one day we decided to outsource some work to a
company called 123-it (http://www.123-it.net) and they use MySQL and I
couldn't believe how much faster transactions were, and with a little
investigation I found that I could make my existing oracle driven
applications work more than 10 times faster using MySQL - has anyone
else found MySQL to be this fast ?

[Olav Tollefsen]

Post by dozey dude
I have been working with Oracle now for almost 10 years and I
obviously think it's probably the best and quickest database system
out there. But then one day we decided to outsource some work to a
company called 123-it (http://www.123-it.net) and they use MySQL and I
couldn't believe how much faster transactions were, and with a little
investigation I found that I could make my existing oracle driven
applications work more than 10 times faster using MySQL - has anyone
else found MySQL to be this fast ?

I have found it to about the same speed as Microsoft SQL Server when you use
a few hours to manually define indexes. Microsoft SQL Server have tools to
analyze and tune indexes. Often database performance is all about using the
optimal indexes. At the end I was unable to use mySQL since it doesn't
support subqueries yet.

Olav

[Christopher Browne]

Post by dozey dude
I have been working with Oracle now for almost 10 years and I
obviously think it's probably the best and quickest database system
out there. But then one day we decided to outsource some work to a
company called 123-it (http://www.123-it.net) and they use MySQL and
I couldn't believe how much faster transactions were, and with a
little investigation I found that I could make my existing oracle
driven applications work more than 10 times faster using MySQL - has
anyone else found MySQL to be this fast ?

Are you sure that it is doing something comparable?

Do some transaction rollbacks on some big updates, and see how long
_that_ takes...

BEGIN;
UPDATE some_big_table SET some_field = 25;
ROLLBACK;

And have you got more than one user doing updates? It hasn't a
history of playing too well under conditions of high concurrency...