Re: [HACKERS] Slow - grindingly slow - query - Mailing list pgsql-hackers

From Theo Kramer
Subject Re: [HACKERS] Slow - grindingly slow - query
Date
Msg-id 382D359A.DB093E91@flame.co.za
Whole thread Raw
In response to Re: [HACKERS] Slow - grindingly slow - query  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Tom Lane wrote:

> The Informix EXPLAIN results that Theo Kramer posted (a few messages
> back in this thread) are pretty interesting too.  If I'm reading that
> printout right, Informix is not any smarter than we are about choosing
> the scan types for the outer and inner queries; and yet they have a much
> faster runtime for the WHERE IN query.

The informix EXPLAIN for the 'not in' query was when I did not have an
index on registrationtype (the explain appends to file sqexplain.out so I
missed it :(). Anyway here is the Informix EXPLAIN with the index on
registrationtype.


QUERY:
------
select accountdetail.domain from accountdetail where accountdetail.domain not in (select accountmaster.domain from
accountmaster)

Estimated Cost: 4510
Estimated # of Rows Returned: 58810

1) informix.accounts: SEQUENTIAL SCAN
   Filters: (informix.accounts.domain != ALL <subquery> AND
informix.accounts.registrationtype != 'N' )
   Subquery:   ---------   Estimated Cost: 12   Estimated # of Rows Returned: 10
   1) informix.accounts: INDEX PATH
       (1) Index Keys: registrationtype           Lower Index Filter: informix.accounts.registrationtype = 'N'


The speed difference with or without the subquery index is neglible for
Informix.
--------
Regards
Theo


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: [HACKERS] Backend build fails in current
Next
From: Peter Eisentraut
Date:
Subject: Re: [HACKERS] Backend build fails in current