To: Robert Haas <robertmhaas@gmail.com>
cc: =?KOI8-U?B?96bUwcymyiD0yc3eydvJzg==?= <tivv00@gmail.com>,
	Ivan Voras <ivoras@freebsd.org>, Jignesh Shah <jkshah@gmail.com>,
	pgsql-performance@postgresql.org
Subject: Re: Performance under contention 
In-reply-to: <AANLkTin=vVgkpYe2u_WBVHV59dgbe5QZ6PZ-nsT28Yb5@mail.gmail.com> 
References: <icccng$kpg$1@dough.gmane.org>
	<AANLkTinFqxaUnpq9E+oh6KcoyqPHDMxjS-_g3AJSPxmk@mail.gmail.com>
	<AANLkTin7iSm3abU8cjWJYJZ4fDuz4Sg69rv17Xh5Z3J1@mail.gmail.com>
	<AANLkTimRPXTs0DXNLQT1nubeNT+tH2JHVXg1-KidvBcH@mail.gmail.com>
	<AANLkTim1EphkUMV=PWV0mF8HaGHqH2H+whnFtqPOnTN_@mail.gmail.com>
	<AANLkTi=q6EFaTuWU9gVFC3g8+6FucFguwBxcyauwEdjh@mail.gmail.com>
	<AANLkTinffMekisiwq9Dip9uWxG2_qeMgPkctw=nCkZmv@mail.gmail.com>
	<AANLkTim2nr+pKJ6ZwduEthzdV94=Ovh9ys_E6nEZCWbr@mail.gmail.com>
	<AANLkTin=vVgkpYe2u_WBVHV59dgbe5QZ6PZ-nsT28Yb5@mail.gmail.com>
Comments: In-reply-to Robert Haas <robertmhaas@gmail.com>
	message dated "Tue, 07 Dec 2010 23:24:14 -0500"
Date: Wed, 08 Dec 2010 09:34:23 -0500
Message-ID: <14229.1291818863@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Robert Haas <robertmhaas@gmail.com> writes:
>> Yeah, that was my concern, too, though Tom seems skeptical (perhaps
>> rightly). šAnd I'm not really sure why the PROCLOCKs need to be in a
>> hash table anyway - if we know the PROC and LOCK we can surely look up
>> the PROCLOCK pretty expensively by following the PROC SHM_QUEUE.

> Err, pretty INexpensively.

There are plenty of scenarios in which a proc might hold hundreds or
even thousands of locks.  pg_dump, for example.  You do not want to be
doing seq search there.

Now, it's possible that you could avoid *ever* needing to search for a
specific PROCLOCK, in which case eliminating the hash calculation
overhead might be worth it.  Of course, you'd still have to replicate
all the space-management functionality of a shared hash table.

			regards, tom lane