Lukas Smith wrote:
mysql: http://marc.theaimsgroup.com/?l=pear-dev&m=108239153527834&w=2
ibase: http://marc.theaimsgroup.com/?l=pear-dev&m=108240351804395&w=2
there are other interesting points being discussed through out this thread but with a noteably focus on the pear abstraction layers.
Interesting thing from my point of view - both done on Apache2
ADOdb is not giving me any problems, but this does show that a switch to raw ibase will give me a performance boost if I need it.
Well raw will always be faster yes.
However take this test with a grain of salt .. its just a few selects with tons of fetches.
Its a very specific situation that is being benchmarked. Things will change even more if you are hitting your memory limited. Than using unbuffered queries will give a huge boost over those abstraction layers that dont support unbuffered queries etc.
regards, Lukas
-- PHP Database Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php