Paged BTree node - Mailing list pgsql-hackers

From Jean Morissette
Subject Paged BTree node
Date
Msg-id 405D05C6.1000800@videotron.ca
Whole thread Raw
List pgsql-hackers
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi developpers,
I would like to know information about OS virtual memory page and your
BTree node.  Theoriticaly, BTree perform better if his size node is the
same that OS page (the unit of transfer between main memory and disk).
This way, it take just one disk access to load in memory a BTree node.
Do you use this strategy for your BTree node?  If yes, could you explain
to me how it works please?  How can you be sure that reading a node in
memory will take just one disk access?  Do you make special call to OS
file system?
Thanks for your time.
Jean
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFAXQXCS/RlGrPlCBoRArJjAKCIg+xAgbyfghqG3J8yTDXq5jGTggCeI7It
v0aH+1pEpue9cxfSqL9Q8cw=
=zBUy
-----END PGP SIGNATURE-----



pgsql-hackers by date:

Previous
From: Fabien COELHO
Date:
Subject: Re: Syntax error reporting (was Re: [PATCHES] syntax error position
Next
From: Yusuf Goolamabbas
Date:
Subject: Why O_SYNC is faster than fsync on ext3