Harald Armin Massa wrote:
> reading postgres benchmarks for beginners advises to stop reading on the
> words "default (ie. unchanged postgresql.conf); but the real test is
> given right after:
>
> http://www.kaltenbrunner.cc/blog/index.php?/archives/21-guid.html
>
> That confirmes my first impression (on different workload) of "the speed
> has doubled".
>
> If reality confirmes, that 8.2 to 8.3 will be a small step in versions,
> and a great step in databases.
yeah but keep in mind that this one only tested a very specific scenario
(update heavy, workload fits easily in buffercache and benefits from
HOT) - it is a fairly neat improvement though ...
Stefan