Re: 1 Billion 11 Byte Words... Need to Check Uniqueness Using Oracle

From: Bryan W. Taylor <bryan_w_taylor_at_yahoo.com>
Date: 12 Feb 2002 14:40:29 -0800
Message-ID: <11d78c87.0202121440.7c3df1d0@posting.google.com>

Mark Townsend <markbtownsend_at_attbi.com> wrote in message news:<B889EF09.10C1E%markbtownsend_at_attbi.com>...

> All the suggestions to date will work, and will work well. However, when I
> first saw the question, I couldn't help wondering if using a database seemed
> to be a bit of overkill. Is this a one of occurence ? Do you need to query
> these words once you have validated uniqueness ? Surely some sort of file
> based sort program would be a lot cheaper if you don't.

Loading and cleansing of large data sets is a routine datawarehouse basic operation, so I'm sure that the original poster proposed this test as a synthetic benchmark to give some sort of measure of how well each RDBMS would handle such work.

It's also not overkill at all if you include development time as a measure of each proposed solution. Received on Tue Feb 12 2002 - 16:40:29 CST