8.2 is 30% better in pgbench than 8.3 - Mailing list pgsql-hackers

From Pavel Stehule
Subject 8.2 is 30% better in pgbench than 8.3
Date
Msg-id 162867790707210840x29d2312ue0925351daed63b5@mail.gmail.com
Whole thread Raw
Responses Re: 8.2 is 30% better in pgbench than 8.3
List pgsql-hackers
Hello

I was little bit surprised. Is any reason for it?

Today I got:

(1) pgbench -c 30 -t 100
(2) pgbench -c 60 -t 100
(3) pgbench -c 30 -t 200
(4) pgbench -c 60 -t 200

autovacuum on
linux (fedora 7, ext3) 1GBRam, Pentium1.6M

Results 8.3 (result 8.2)

1.  300/303 (526/536)
3.  259/261 (423/429)
3.  310/312 (529/534)
4.  240/241 (413/416)

Regards
Pavel Stehule


pgsql-hackers by date:

Previous
From: "Mickael DELOISON"
Date:
Subject: Query plan and execution time of a query
Next
From: Tom Lane
Date:
Subject: Re: 8.2 is 30% better in pgbench than 8.3