Обсуждение: Distributed / Linked with Open Source DB

Поиск
Список
Период
Сортировка

Distributed / Linked with Open Source DB

От
"Ex Fed"
Дата:
Does anybody out there have experience implementing either a distributed
homogenous database or linked databases with MYSQL or POSTGRES.

The reason I ask is I will have 4 distant regions connected over high speed
data lines, that will have seperate databases, if the databases had the same
table structure, ie table names and field types and widths the same, would
it easily be possible to link these together so that they can be queried as
one database (union), but the one region can only perform DML on their own
databases data.  Has anybody out there done this?

I assume this would require a lot of traffic between the servers.
Is this efficient, and reliable?

I am looking at Oracle, they offer all of the capabilities, but are pretty
complex, and way more expensive than the open source solutions.


Lee D.



_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com


Re: Distributed / Linked with Open Source DB

От
"Shridhar Daithankar"
Дата:
On 4 Oct 2002 at 14:49, Ex Fed wrote:

> Does anybody out there have experience implementing either a distributed
> homogenous database or linked databases with MYSQL or POSTGRES.
>
> The reason I ask is I will have 4 distant regions connected over high speed
> data lines, that will have seperate databases, if the databases had the same
> table structure, ie table names and field types and widths the same, would
> it easily be possible to link these together so that they can be queried as
> one database (union), but the one region can only perform DML on their own
> databases data.  Has anybody out there done this?
>
> I assume this would require a lot of traffic between the servers.
> Is this efficient, and reliable?

If you implement some kind of replication(Plenty of it available for postgres)
which periodically syncs all the databases, you can query individual nodes and
get pretty accurate picture.

I hope you don't need real time replication here. It's available for postgresql
but would be too much bandwidth intensive..

Bye
 Shridhar

--
My father taught me three things:    (1) Never mix whiskey with anything but
water.    (2) Never try to draw to an inside straight.    (3) Never discuss business
with anyone who refuses to give his name.