Re: Running 9 in production? Sticking with 8.4.4 for a while?

From: Francisco Reyes <lists(at)stringsutils(dot)com>
To: Tory M Blue <tmblue(at)gmail(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Running 9 in production? Sticking with 8.4.4 for a while?
Date: 2010-10-27 15:19:43
Message-ID: cone.1288192783.584948.5695.1000@shelca
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Tory M Blue writes:

> Just looking for 9.0 feedback (understand it's still quite new).

Although not 9.0 feedback per se.. but something to consider...

I am moving a number of machines from 8.1 to 8.4.
We use a rolling procedure with a spare:
machine A running 8.1
machine B is prepped with 8.4
change names and IPs
keep old A for a bit just in case...

The last replacement we noticed a big regression in performance. Upon
inspection the machine we replaced was 3.7.. Both 8.X and 8.X have serious
problems with a particualr query (dealing with it on performance now).

So I guess what I am trying to say... you need to test with your own data
set. That lone 8.3.7 machine was literally a fluke. We were going to move
all 8.1 to 8.4.. but at one point did one in 8.3. If we had not done that we
would not have noticed that both 8.1 and 8.4 have problems with some of our
queries.

I am even leaning now to have a lab with 8.3 for testing queries.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Steve Singer 2010-10-27 15:42:09 Re: Postgres insert performance and storage requirement compared to Oracle
Previous Message Francisco Reyes 2010-10-27 15:14:05 Re: Regression: 8.3 2 seconds -> 8.4 100+ seconds