Re: PostgreSQL Replication and more

From: "David Siebert" <david(at)eclipsecat(dot)com>
To: <pgsql-general(at)postgresql(dot)org>
Subject: Re: PostgreSQL Replication and more
Date: 2002-01-14 20:13:19
Message-ID: OJEIJALIHAIBMMBFLCOBOENIDMAA.david@eclipsecat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


I have been please with the Query speed on Postgres. I am running postgres
on a really old machine 100 mhz Pentium with only 32 megs of ram and only 32
megs of ram. My app spends a lot of time running querys. from the speed and
CPU load I am getting I think I could run about 5o users on it.
I have heard that Postgres is slower than MySQL for inserts in version 7.0
and below.
I can only give you a half answer Query's seem fast to me. Inserts are not
an issue for me.

-----Original Message-----
From: pgsql-general-owner(at)postgresql(dot)org
[mailto:pgsql-general-owner(at)postgresql(dot)org]On Behalf Of Jeppe S
Sent: Tuesday, January 08, 2002 1:02 PM
To: pgsql-general(at)postgresql(dot)org
Subject: [GENERAL] PostgreSQL Replication and more

I am considering jumping ships in the great scheme of open-source databases.
After running into "trouble" in mysql - it doesn't optimize the use of "OR"
in queries, amongst other concerns, I am considering porting to PostgreSQL.

However, a few things worry me.

1. Replication

I see that PostgreSQL doesn't natively support replication. However, I've
found rserv and pgreplicate. Both look like decent projects, but I'm curious
if anyone here has toyed with them at any length or used them in a
production environment. I will be needing production quality replication.

2. Performance (particularily many small queries and inserts)

Due to the nature of the app I work on, and historical reasons as well, I
perform many small queries that need to go quite quickly. I rewrote some
code to use one big query and the speed increased by a factor of 10.
However, since mysql lacks subselects, I haven't been able to rid myself of
the many-small-queries approach (specifically, i'm getting transaction
groups with - say - 2 day's history, but from within an initial time window
of 3 hours). This can be done with subselects, but i am curious about the
performance of subselects.

Also, what is the insert performance of PostgreSQL? Compared to mysql? I'm
doing transaction processing, and receive up to ten transactions per second.
Don't worry, I can throw hardware at the problem, but I'd rather not.

I hope some of you got some experiences to share with me, positive or
negative.

Best regards,

Jeppe

---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to majordomo(at)postgresql(dot)org)

In response to

Browse pgsql-general by date

  From Date Subject
Next Message David Siebert 2002-01-14 20:13:20 Stored proceures in perl or Python
Previous Message wsheldah 2002-01-14 20:06:45 Re: Integrating ADSM