SV: MySQL and PostgreSQL speed compare - Mailing list pgsql-general

Hi,

...
> > I wrote a piece of benchmarking, just to test my classes, and
> > was suprised of the speed diffs.
...
> The fact that MySQL doesn't support transactions at all severly
> limits its utility for applications that need data consistancy, it

I thought it does. Well I need to test that before Im going to say anything
more.

...
> A lot of things went wrong here, first off you didn't contact the
> developers to let them know ahead of time and discuss tuning the
> system.  Both the MySQL and Postgresql developers deserve a chance
> to recommend tuneing for your application/bench or ask that you
> delay your bench until bug X or Y is addressed.

I did not (and do not) take it so seriously. I do not (did not) claim that
the test is in any way usefull I just wanted peoples response (and did get
some).

...
> Most sites are that I know of are dynamic content and perform
> selects for the most part.

Yes but still.

...
> You have an admitted inbalance with the disk systems but don't go
> into any details.

Yes that was sloppy. I thought of that after I started to write my email. I
did a fast test with both dbas on SCSI (just simple moved PostgreSQL "data"
directory to SCSI). But the result was even slower. Anyway the biggest
differece between SCSI and IDE is throughput and CPU usage. Throughput is
not an issue here (small chunks of data), and CPU should not be (PostgreSQL
peaked CPU with 20 connections on IDE and used 75% on SCSI). It might be a
bigger difference with more connections. Both my IDE and SCSI are quite new
with fast search.

...
> You probably didn't tune postgresql worth a damn.  I don't see any

Neither did I tune MySQL. Neither do 90% of the users.

...
> mention of you raising the amount of shared memory allocated to
> postgresql.  I also imagine you may have run the test many times
> on Postgresql without vacuuming the database?

The test program DROPS the tables and recreates them. I do not know if you
still would need to VACUUM the dba. Anyway I did run the test several times
wihtout seing any (big) differences.

...
> I think your time would be better spent working on actually
> impelementing the features you want rather than posting broken and
> biased benchmarks that do more harm than good.

I do not think this was biased, maybe broken but not biased. Actually I use
PostgreSQL and all (free?) dbas that Ive installed have been PostgreSQL. The
code Im using was first written for PostgreSQL and the reason why I added
MySQL support was that my ISP refused to install PostgreSQL. I did the test
just to see if my classes also worked on MySQL before starting to port rest
of my code to MySQL (guess if I was suprised).

Implementing features... why? PostgreSQL has (almost) everything I need. Its
MySQL which would need to have some new features (sub-selects, views...).

Looking for the truth and nothing but the truth? Dont look for it among
humans. All they have are opinions.

// Jarmo