Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Mailing Lists -> Oracle-L -> RE: Oracle RAC cost justification?
Has anyone implemented 'true TAF' i.e. TAF that works with updates, deletes and inserts as opposed to just selects. I know it is not supported but there are ways around it. Has anyone done it. I don't see how a business can do just selects and be OK justifying the cost of RAC.
Does anyone have an application that does JUST SELECTS then it will be a true Transparent Application Failover - does anyone ? Raj, can you maybe discuss a little bit about the way your application works ?
Tapan Trivedi
David <thump_at_cosmiccooler.org> wrote:
It works fine...
Users get errors and have to reconnect.
If that is okay, then that is cool.
For what I do though, that is unacceptable.
My point is I bet you can accomplish the same without RAC.
-----Original Message-----
From: oracle-l-bounce_at_freelists.org =
[mailto:oracle-l-bounce_at_freelists.org]
On Behalf Of rjamya
Sent: Sunday, June 05, 2005 5:55 AM
To: thump_at_cosmiccooler.org
Cc: mgogala_at_allegientsystems.com; tim_at_evdbt.com; Oracle Discussion List
Subject: Re: Oracle RAC cost justification?
We do and I can tell you, if implemeted correctly, it works fine. =
Hundreds=20
of our users who rely on split-second response times do not even know if =
their application gets moved from one node to another. Might be because =
for=20
such critical sessions we do pre-connects. but yes, it works. For=20
applications it is usually few seconds when they get errors and have to=20
connect again.
Raj
On 6/2/05, David wrote:
>=20
> the others heads in terms of smon/eviction? How long does it take to=20
> get the application to start hitting an instance an another node?
>=20
>=20
-- http://www.freelists.org/webpage/oracle-l -- http://www.freelists.org/webpage/oracle-l Tapan Trivedi Senior Oracle DBA 916 613 1921 test'; "> test'; "> -- http://www.freelists.org/webpage/oracle-lReceived on Sun Jun 05 2005 - 16:25:16 CDT
![]() |
![]() |