High Availability - Performace Scalability - Dyanamic Load Balancing

Jerome,

I'm curious what you expect out of clustering. High Availability, and
Load Balancing seem possible, but Performance ? I would think there is a
cost associated with clustering, that degrades performance.

Anyways, to answer your question, there are a few tools that allow High
Availability, and Load Balancing.

pgpool ftp://ftp.sra.co.jp/pub/cmd/postgres/pgpool/pgpool-2.0.8.tar.gz

and if you prefer java C-JDBC
http://c-jdbc.objectweb.org/

Dave
On Wed, 2004-08-18 at 06:10, JM wrote:

Hi All,

I just came off an exibit and I was wondering if there ways to do clustering
in postgres.
I envy mySQL because there's a application for clustering it...

EMIC networks has this marketing tagline:
High Availability - Performace Scalability - Dyanamic Load Balancing
they are the one who provide this app... ( but its not open source )

Thanks,

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

--
Dave Cramer
519 939 0336
ICQ # 14675561

Dave Cramer wrote:

Jerome,

I'm curious what you expect out of clustering. High Availability, and
Load Balancing seem possible, but Performance ? I would think there is a
cost associated with clustering, that degrades performance.

Anyways, to answer your question, there are a few tools that allow High
Availability, and Load Balancing.

pgpool ftp://ftp.sra.co.jp/pub/cmd/postgres/pgpool/pgpool-2.0.8.tar.gz

and if you prefer java C-JDBC
http://c-jdbc.objectweb.org/

Another interesting tool is SQLRelay:

http://sqlrelay.sf.net/

-- Mitch

On 2004-08-18, Dave Cramer <pg@fastcrypt.com> wrote:

Jerome,

I'm curious what you expect out of clustering. High Availability, and
Load Balancing seem possible, but Performance ? I would think there is a
cost associated with clustering, that degrades performance.

Data in an RDBMS application should be localised enough that you
should be able to evenly split the workload over a number of cluster
nodes and not have them step on each other's toes too much. It's a difficult
problem but not insurmountable.

[deletia]

--
Negligence will never equal intent, no matter how you
attempt to distort reality to do so. This is what separates |||
the real butchers from average Joes (or Fritzes) caught up in / | \
events not in their control.