DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma;
	h=message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition;
	b=I+RDUossgGoS7PlIj4IlWutQBLxbCyqVcoT3e4yEst12IdxR8RdLEASkLsdkw/ouPhGpunUrlot/thAoztuJy9Eop3sIS/qN3yMEMNIkDXceLZmx158it0v3Nf87Pe1dfwGMjug3S9Ijcv5CJeomV/GzO49tZgmfbDzv2aEuvus=
Message-ID: <c0787d8a0804100748l51b63c7fx34db59e33f13c208@mail.gmail.com>
Date: Thu, 10 Apr 2008 10:48:59 -0400
From: "Jon Stewart" <jonathan.l.stewart@gmail.com>
To: pgsql-performance@postgresql.org
Subject: Creating large database of MD5 hash values
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Hello,

I am creating a large database of MD5 hash values. I am a relative
newb with PostgreSQL (or any database for that matter). The schema and
operation will be quite simple -- only a few tables, probably no
stored procedures -- but I may easily end up with several hundred
million rows of hash values, possible even get into the billions. The
hash values will be organized into logical sets, with a many-many
relationship. I have some questions before I set out on this endeavor,
however, and would appreciate any and all feedback, including SWAGs,
WAGs, and outright lies. :-) I am trying to batch up operations as
much as possible, so I will largely be doing comparisons of whole
sets, with bulk COPY importing. I hope to avoid single hash value
lookup as much as possible.

1. Which datatype should I use to represent the hash value? UUIDs are
also 16 bytes...
2. Does it make sense to denormalize the hash set relationships?
3. Should I index?
4. What other data structure options would it make sense for me to choose?

Thanks in advance,


Jon