Re: Very fast select

There are only a limited number of things you can do..

Here are a few suggestions off the top of my head...

1. Work out how big the total table size is and see if it is possible to pin the table into the SGA. I suspect that this is not feasible.
2. Ensure that you have few defragmented extents on the table and associated indexes.
3. Increase the db_file_multiblock_read_count parameter so that more data can be read in on each disk read.
4. Consider the possibility of using Dedicated server if you are already using Multi-Threaded
5. Put the table on a seperate tablespace and fast disk unused by anything else.

Any other ideas anyone ?

<iustinami_at_my-deja.com> wrote in message news:95eu2r$ci2$1_at_nnrp1.deja.com...
> Hi,
>
> I need very fast answers to <order by> selects
> on a table ( max 1-2 seconds ) that must contain
> a varchar field ( the <order by> field ). The
> table has about 300000 rows.
> I've tried indexing, primary key on varchar
> field, but still too slow. The server is Oracle
> 8i on a K7, 256Mb running Win2000 ).
> I think I need something like the low-level
> order of rows in database to be exactly the
> varchar field order.
> If you have any ideea please help.
>
> Thanks in advance,
> Iustin Amihaesei
>
>
>
>
> Sent via Deja.com
> http://www.deja.com/
Received on Sat Feb 03 2001 - 12:22:07 CST