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: Avoiding generating redo logs

Re: Avoiding generating redo logs

From: Jeremy <jeremy0505_at_gmail.com>
Date: Thu, 21 Sep 2006 08:34:11 +0100
Message-ID: <MPG.1f7c50d2ff9d9b1f98a2cd@news.individual.net>


In article <u3bal4umo.fsf_at_rcn.com>, Galen Boyer says...
>
> But, then, when the user subsequently looks at the data, it is not in
> real-time. It, instead, is at some time after the "snapshot" was
> taken. How come they are okay with the subsequent reads of the data,
> but are not okay with the data maybe, already, being a little behind
> exact committed transactions at the time of the initial query?
>
>

Well the first time the user clicks to see the data, it will run the query and present the results. For sure, when he changes the sort order by clicking a column heading, this is going to read the data from the "cache" table. As long as the user knows that the data was 100% up-to- date at the time of the intial query, that is entirely adequate.

-- 

jeremy
Received on Thu Sep 21 2006 - 02:34:11 CDT

Original text of this message

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