From: dyou98@aol.com (D.Y.)
Newsgroups: comp.databases.oracle
Subject: Re: performance question
Date: 6 Jun 2002 08:21:07 -0700
Organization: http://groups.google.com/
Lines: 28
Message-ID: <f369a0eb.0206060721.450ec53b@posting.google.com>
References: <16b21687.0206051302.7a662829@posting.google.com>
NNTP-Posting-Host: 208.28.120.74
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1023376868 15585 127.0.0.1 (6 Jun 2002 15:21:08 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: 6 Jun 2002 15:21:08 GMT


Sounds like you have a DSS/OLAP type of database. Read up on materialized
view and query rewrite. It may help you.

Vissuyk@yahoo.com (Vissu) wrote in message news:<16b21687.0206051302.7a662829@posting.google.com>...
> Hi All,
> 
> We have multiple tables with over 15 million rows and growing fast. 
> There is a about 50+ columns in these tables.
> 
> Some queries use max(date) function to figure out the latest and
> greatest value belonging to a certain group. Lot of queries do order
> by and group by.
> 
> Some queries are fast and some are slow. I want to recreate these
> tables and with the goal of making it fast and scalable.
> 
> I am wondering if I should index organize these tables. I am not sure
> if I can use IOTs for such volume of data. Alternatively, I am
> thinking of partitions etc.
> 
> Could anyone suggest better ways to achieve scalability and
> perfromance?
> 
> RDBMS = Oracle 901
> 
> Thanks much,
> 
> Vissu

