Re: partioning option not worth it?

I don't depend on partition views, merely support for union all in a view which should be indefinite.

The effort doesn't change dependent on the number of developers if you get the architecture right - it's just a matter of getting the table name to use before running the query, which I'm doing anyway. This approach is easy enough to implement in most development tools (vb, delphi, c++, powerbuilder). I don't know about developer (I gave up using forms when version 4 came out). Other tools can be a pain in arse to modify this way (crystal reports comes to mind, but that's a pain in the arse anyway).

I agree that the cost/benefit analysis can depend on the number of partitioned tables, but in most systems using partitioning I would expect there to be a relatively small number.
Because the additional charge for partitioning depends on the number of servers you have, both development effort for the alternative and the cost of parititioning increase as the number development teams/developers increases.

In my case, the additional coding cost in the front-end is very low since the program dynamically builds queries anyway. The biggest pain will be updating the batch job to populate N tables instead of 1 - fortunately, I don't need to parallelise processing to get adequate throughput.

Again, I am more than happy to admit that there may be times when the partitioning option / ee are essential or desirable. I'm merely contending that most of the time it is not.

There is also the issue that in many large organisations the offering to the great god Oracle has already been made so it's irrelevant.

John Darrah <jdarrah_at_veripost.net> wrote in message news:6875fa94347268d8f18b5986459c0a6c.36240_at_mygate.mailgate.org...
> I think you are kidding yourself with your one day estimate but mabey
> not. Having gone down the road of creating "roll your own" workarounds
> to existing functionallity, all I can say is yes, its possible but in a
> lot of environments, its impractical. With one developer, the
> "the pain in the ass factor" of splitting up tables into partition views
> is 2 (double that of using partitioning). The problem is that this
> number goes up exponentially the more developers / groups of developers
> you have working in your company. Also, I'm pretty sure that Oracle is
> desupporting partition views in release 9.2 (anyone have confirmation on
> that one? I'm too lazy to look it up right now) in which case your
> solution has a limited life span. Not saying its a bad idea just that
> it is probably not as easy as you make it seem.
>
>
> --
> Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
Received on Fri Feb 22 2002 - 02:37:24 CST