Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Real World Experience of Oracle RAC and its ability to Scale

Re: Real World Experience of Oracle RAC and its ability to Scale

From: DA Morgan <damorgan_at_x.washington.edu>
Date: Wed, 20 Apr 2005 23:10:59 -0700
Message-ID: <1114063623.882755@yasure>


Ian Turner wrote:

> When you talk about >4 node Oracle RAC clusters are you talking database
> server - or application and database server.
>
> I understand a little more about the application architecture being proposed
> now.
>
> They are attempting to create a grid architecture to manage both the
> application and database components. This is being done using Oracle CRS.
> I am being told that for CRS to be able to manage a node it must have a
> database instance on it and be part of the RAC cluster. However, unlike the
> database instance the application components are not load balanced. The
> only cluser management of application components appears to be to restart
> application components in the event of one of them failing. CRS must
> perform some monitoring to support this.
>
> This seems odd to me. I would have thought OEM could perform this function,
> e.g. install the Oracle agent on the application server, configure it to
> report some custom events and create scripts/triggers within OEM to restart
> failed components. I actually thought OEM managed the grid environment - I
> wasn't aware of CRS until today.
>
> Any thoughts appreciated,
>
> Ian

Assuming 10g I don't believe this is correct.

Contact me off-line and I will put you in touch with someone that is an expert on the subject.

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu
(replace 'x' with 'u' to respond)
Received on Thu Apr 21 2005 - 01:10:59 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US