Re: BIG Joins

In article <302ahl$j4p_at_tamsun.tamu.edu> m0v5533_at_tamsun.tamu.edu (Mahesh Vallampati) writes:
>From: m0v5533_at_tamsun.tamu.edu (Mahesh Vallampati)
>Subject: Re: BIG Joins
>Date: 13 Jul 1994 22:13:57 -0500
 

>In article <9407131738.PN05420_at_ll.mit.edu>,
>James Forgy <forgy_at_ll.mit.edu> wrote:
>>In article <Csw32B.Ju7_at_Unify.com> jde_at_Unify.com (Jeff Evarts) writes:
>>>From: jde_at_Unify.com (Jeff Evarts)
>>>Subject: BIG Joins
>>>Date: Wed, 13 Jul 1994 17:25:23 GMT
>>Of course I have written them. Sometimes you have to join a table to itself
>>and then suddenly are faced with maybe three look-up tables and some outer-join
>>logic that involves two more tables.
>>I have wrote a 7-Way Join with a Union on a 600,000 row table and it only took
>>7 seconds to run.
>Wow,can u provide more details like which machine used and the number of row
>in the seven tables u joined.Again if u had used a selection criteria on the
>600,000 row table which u use to union with the joined table then the above
>claim would be valid because the query optimiser would be smart enough to
>filter out the rows in the big table. As it stands i think the above is a
>tall claim.I would appreciate it if u could give more details on the above.
>>-jf
>>>I`m summarize to these newsgroups.
>>>-Jeff Evarts
>>>--jde_at_unify.com
 

>Thanks and Regards
>Mahesh Vallampati
>M.S. In EE
>Dept.of Electrical Engineering,
>Texas A & M University.
>Ph:(409)845-6189
>\\ In the Beginning there was Codd.....

That's not a claim buddy, that's called "set timing on", mabye I will download this SQL from an HP9000 and post it for you.

-jf Received on Mon Jul 18 1994 - 18:14:30 CEST