Re: TPC-C and Postgres - Mailing list pgsql-general

From Mark Kirkwood
Subject Re: TPC-C and Postgres
Date
Msg-id 3FF7BD60.9000104@paradise.net.nz
Whole thread Raw
In response to Re: TPC-C and Postgres  ("Keith C. Perry" <netadmin@vcsn.com>)
Responses Re: TPC-C and Postgres
List pgsql-general


>On page 7 of this document:
>
>"Earlier versions of Postgres were often unstable and quite often there were
>even data losses. The Postgres developer team has invested a great deal of time
>in regression tests, which guarantee high stability and data security. Releases
>now only occur after longer Beta phases. The new versions of Postgres have
>become considerably more effective (see box  Benchmarks ).
>
>There are not known to be any serious stability problems with MySQL. With the
>new feature of replicability, the system has taken another major step in the
>direction of increased failure safety. Especially in conjunction with PHP, MySQL
>achieves good performance values in web applications (see box  Benchmarks ). A
>paper by the founder of MySQL, Michael Widenius on performance optimisation is
>recommended in this respect"
>
>
>Hmmmm...
>
>
>

Sounds like he has Mysql And Pg confused... :-)

e.g :

Recently a colleage of mine, who is fond of Mysql initiated a friendly
challenge to the effect "Mysql can do big queries just as well as Pg".

I suggested he try my Data Warehouse benchmark
(http://sourceforge.net/projects/benchw)
with size 10 (about 10G or so).

I went away, produced my results (using Pg 7.4) and mailed them to
him... however at his end - he cannot get Mysql to load the data (it
kills his connection,.. thats nice). He is not impressed at all, and is
considering using Pg!

(he was using Mysql 4.0.15)

regards

Mark





pgsql-general by date:

Previous
From: "Keith C. Perry"
Date:
Subject: Re: TPC-C and Postgres
Next
From: Mark Kirkwood
Date:
Subject: Re: why the need for is null?