Re: performance impact of index

From: Daniel Morgan <damorgan_at_x.washington.edu>
Date: Wed, 26 May 2004 23:11:56 -0700
Message-ID: <1085638322.274883_at_yasure>


Bricklen wrote:
> mrFarenheit wrote:
>

>>>
>> I your schema is purely read-only, you can create as much indexes as 
>> you want. But bitmapped indexes are only useful when the cardinality 
>> is relatively low (about 1.000 distinct values max).
>> In other cases (more than one thousand distinct values for the 
>> column), it can be better to use b-tree indexes, but don't forget that 
>> the storage needed is much bigger.
>> You can alos deal with function-based indexes if you want.
>> .

>
>
> Would you mind proving your statement "bitmapped indexes are only useful
> when the cardinality is relatively low (about 1.000 distinct values max)"?
>
> I'd be interested to see how you came up with that number, and the
> ratios you've used for it.

I would too. From everything I've read and done I'd use a number more closer to 10.

-- 
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)
Received on Thu May 27 2004 - 08:11:56 CEST

Original text of this message