Re: 8.3devel slower than 8.2 under read-only load

"Guillaume Smet" <guillaume.smet@gmail.com> writes:

> On Nov 24, 2007 5:16 PM, Gregory Stark <stark@enterprisedb.com> wrote:
>> Several of the major changes in 8.3 are I/O vs CPU tradeoffs which could be
>> causing a slowdown if you're measuring primarily CPU resources. I'm thinking
>> of both HOT and packed varlenas. I don't know if either of these are causing
>> your slowdown but it's possible.
>
> Here are some rough results:
> http://people.openwide.fr/~gsmet/postgresql/postgresql_8.3_development_cycle_1.png
>
> I don't pretend that this bench is realistic but it's a first step. I
> run the tests with weird numbers more than 3 times to check that they
> are consistent.
>
> IIRC, packed varlenas were commited during April and HOT at the end of
> September.

Hm, so your test seems to be showing up a difference due to packed varlena?
What functions or operators are you using? Could you send the schema and
actual queries?

--  Gregory Stark EnterpriseDB          http://www.enterprisedb.com Get trained by Bruce Momjian - ask me about
EnterpriseDB'sPostgreSQL training!