Re: row vs row.column level locking

On Fri, 16 Dec 2005 01:58:57 GMT, Mladen Gogala <gogala_at_sbcglobal.net> wrote:

>On Thu, 15 Dec 2005 19:03:20 +0000, NetComrade wrote:
>
>> I feel different processes should be able to update different columns
>> w/o blocking each other, as long as what's updated is not the PK.
>
>Why do you feel that?

Why not? Let's say i have a table within a web application that has fields that are updatable by end-user and and fields updatable by a batch job. Currently I have to denormalize and separate the two tables, so that user won't block batch and vice versa. If I have multiple batch jobs running at different times I might have to do even more separation (if the batch jobs are taking too long)...

No-one clearly explained to me why not to have such a feature. The main reason I can think of is that Oracle would have to do much more work to track changes for 'read-consistency'

.......
We run Oracle 9.2.0.6 on RH4 AMD
remove NSPAM to email Received on Fri Dec 16 2005 - 13:22:15 CST