[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: [zzdev] Oooops big problem in zigzag implementation
- To: zzdev@xxxxxxxxxx
- Subject: Re: [zzdev] Oooops big problem in zigzag implementation
- From: Andrew Pam <xanni@xxxxxxxxxxxxxxxxx>
- Date: Wed, 18 Nov 1998 17:32:42 +1100
- In-reply-to: <19981029204708.22172.qmail@xxxxxxxxxx>; from Mark-Jason Dominus on Thu, Oct 29, 1998 at 03:47:08PM -0500
- References: <19981029204708.22172.qmail@xxxxxxxxxx>
- Reply-to: zzdev@xxxxxxxxxx
On Thu, Oct 29, 1998 at 03:47:08PM -0500, Mark-Jason Dominus wrote:
> There is no efficient way to find out what links there are from a
> particular cell. You have to grovel over the entire database looking
> for the links.
Well, it's not quite as bad as that; you only have to try every dimension
to see which ones have a link. However once there are a very large number
of dimensions this becomes quite inefficient. One simple solution I have
considered is changing from a hash database to a database with ordered
indexes such as a B-tree (trivial change to use the existing DB database
in B-tree mode instead of hash mode). This trades off the speed of
insertions and retrievals gained with a hash database against improved
speed of finding all the links from a cell. At present the former is
far more common than the latter, so this might be counter-productive.
Cheers,
*** Xanni ***
--
mailto:xanni@xxxxxxxxxx Andrew Pam
http://www.xanadu.com.au/ Technical VP, Xanadu
http://www.glasswings.com.au/ Technical Editor, Glass Wings
http://www.sericyb.com.au/sc/ Manager, Serious Cybernetics
P.O. Box 26, East Melbourne VIC 8002 Australia Phone +61 3 96511511