Re: Loading 300 million records

Anurag Varma wrote:
> Daniel,
>
> Not knowing OP's overall process and considering that OP is doing a
> insert /*+ append */ into maintable select ... from <join of 5
> tables>;
>
> ... the idea of selecting from ext_tab and etl transforming does not
> really compute.
>
> Although, tuning his sql (maybe use parallel processing etc.) might
> speed things up
> for him.
>
> The paper as such is ok. It does have some nice ideas about etl
> processing
> (especially the example you state above).
>
> Cheers,
>
> Anurag

Just to throw in another 0.02 EUR: as the OP seems to be copying data in the same database and we don't have further information about the scenario, it might as well be a solution to replace the table with a view that does the join. Depending on the types of queries ran aganst the large master table proper indexing might be the solution. Of course, this does not apply if the data is deleted from the tables it was read from after the insert...

Kind regards

    robert Received on Tue Jun 07 2005 - 03:39:51 CDT