[wp-hackers] Improving WordPress' Performance WAS Changing MySQL
komra at design4results.com
Thu Nov 30 19:55:40 GMT 2006
My 2c. I have been seriously in the throes of the MT/WP performance
issues for months now. Neither platform really has it nailed. In MT
if you go to static publishing, which is faster for the end user, the
editorial experience is impossibly slow. I finally found and used an
undocumented page caching scheme (built into MT!) to help the overall
scene. Nevertheless, we have bitten the bullet and moved to WP.
On WP, out of the box there are some expensive queries that are
significantly remedied simply by adding indexes to certain tables. We
run one installation under eAccelerator, and one under
phpAccelerator. And after the database optimization, I still plan to
install a hard page caching scheme, since the queries for paged
category archives are the only slow queries that remain. (FWIW, WP-
Cache does not behave well in either install.)
On Nov 30, 2006, at 2:22 PM, Doug Stewart wrote:
> On 11/30/06, Austin Matzko <if.website at gmail.com> wrote:
>> On 11/30/06, Computer Guru <computerguru at neosmart.net> wrote:
>> > No matter how you look at it, WP is horribly inefficient and now
>> > definitely big enough that this warrants a second look. We need
>> to do
>> > something, the only question is, what?
>> I'm not saying you're wrong, but could you quantify this
>> In other words, how are you measuring it especially in comparison
>> with other platforms?
> WordPress performance under load vs. Movable Type (static publishing)
> performance under load. MT whips WP, but it's actually kind of a
> "cheat" - the entries themselves are statically generated and stored.
> Any dynamic elements (comments, plug-ins, etc.) are then generated
> As I stated above, a good metric is the consummate *thrashing*
> out-of-the-box WP sites take when dugg/slashdotted/etc.
> wp-hackers mailing list
> wp-hackers at lists.automattic.com
More information about the wp-hackers