From: | Vadim Mikheev <vadim(at)krs(dot)ru> |
---|---|
To: | Michael Meskes <Michael(dot)Meskes(at)usa(dot)net> |
Cc: | PostgreSQL Hacker <pgsql-hackers(at)postgreSQL(dot)org> |
Subject: | Re: [HACKERS] performance test |
Date: | 1999-01-19 03:27:50 |
Message-ID: | 36A3FBB6.B37475A3@krs.ru |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Michael Meskes wrote:
>
> I tried inserting 15000 tuples each in two tables via pgsql and timed it.
> After insertion I also created btree indeces on both tables.
>
...
>
> Then I switched to 6.5 without -F:
> real 2m35.833s
> user 0m2.900s
> sys 0m3.910s
>
...
>
> Then I switched on -F:
> real 1m11.573s
> user 0m3.150s
> sys 0m3.760s
>
> I've seen even bigger advantages of -F.
You've seen disadvantages of our unperfect buffer manager -:)
When server need in buffer for new data comming and there is
no unused buffers in pool (i.e. - all buffers filled with new
data and marked as dirty), server gets some dirty buffer,
writes it AND FSYNC file. So, server does many fsyncs
even with BEGIN/END while should do _one_ fsync at COMMIT.
Having this problem fixed you wouldn't had so big difference
between -F and BEGIN/END-no-F.
Vadim
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 1999-01-19 05:25:48 | Re: [HACKERS] Postgres Speed or lack thereof |
Previous Message | Bruce Momjian | 1999-01-19 03:00:14 | Re: [HACKERS] latest parser changes |