Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
| From | Andrew Reilly <areilly---@bigpond.net.au> |
|---|---|
| Newsgroups | comp.arch |
| Subject | Re: Hardware for linked lists |
| Date | 2012-03-06 11:38 +0000 |
| Message-ID | <9rmb8uFb3U1@mid.individual.net> (permalink) |
| References | <J9R4r.11081$jU1.4604@newsfe08.iad> <464d29-qcc.ln1@ntp6.tmsw.no> <4F543DC9.4060206@SPAM.comp-arch.net> <4F558A53.90500@SPAM.comp-arch.net> |
On Mon, 05 Mar 2012 19:53:55 -0800, Andy (Super) Glew wrote: > Arrays are only easier for hardware if accessed sequebntially, or in > some other numerically regular pattern. Yes, but (singly, doubly) linked lists demand sequential access, and that doesn't seem to be doing much to stop (a certain class of) programmers using them. Take your order-of-magnitude performance improvement where you can find it! Yes, "big data" database and filesystem coders know other techniques, but they're fairly thin on the ground. Until Steele's "divide by halving, rather than "first | rest" (aka accumulators are poison) dictum, or the notion of abstract collections that understand associativity and commutativity become widespread, code is going to walk through data structures in-order. Sequential access is a good bet, for hardware. IMO. Cheers, -- Andrew
Back to comp.arch | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
Hardware for linked lists Robert Myers <rbmyersusa@gmail.com> - 2012-03-04 16:29 -0500
Re: Hardware for linked lists Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-03-04 22:41 +0100
Re: Hardware for linked lists "Andy (Super) Glew" <andy@SPAM.comp-arch.net> - 2012-03-04 20:15 -0800
Re: Hardware for linked lists "Andy (Super) Glew" <andy@SPAM.comp-arch.net> - 2012-03-05 19:53 -0800
Re: Hardware for linked lists Andrew Reilly <areilly---@bigpond.net.au> - 2012-03-06 11:38 +0000
Re: Hardware for linked lists nmm1@cam.ac.uk - 2012-03-06 12:16 +0000
Re: Hardware for linked lists Bakul Shah <usenet@bitblocks.com> - 2012-03-06 23:24 -0800
Re: Hardware for linked lists Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-03-07 09:08 +0100
Re: Hardware for linked lists Noob <root@127.0.0.1> - 2012-03-06 13:25 +0100
Re: Hardware for linked lists Terje Mathisen <"terje.mathisen at tmsw.no"> - 2012-03-06 17:11 +0100
Re: Hardware for linked lists nmm1@cam.ac.uk - 2012-03-04 21:51 +0000
Re: Hardware for linked lists Robert Myers <rbmyersusa@gmail.com> - 2012-03-04 17:16 -0500
Re: Hardware for linked lists nmm1@cam.ac.uk - 2012-03-04 23:15 +0000
Re: Hardware for linked lists Chris Gray <cg@GraySage.com> - 2012-03-04 17:04 -0700
Re: Hardware for linked lists nmm1@cam.ac.uk - 2012-03-05 09:01 +0000
Re: Hardware for linked lists Chris Gray <cg@GraySage.com> - 2012-03-05 14:07 -0700
Re: Hardware for linked lists George Neuner <gneuner2@comcast.net> - 2012-03-05 00:57 -0500
Re: Hardware for linked lists cjt <cheljuba@prodigy.net> - 2012-03-04 19:09 -0600
Re: Hardware for linked lists MitchAlsup <MitchAlsup@aol.com> - 2012-03-04 18:24 -0800
Re: Hardware for linked lists Robert Myers <rbmyersusa@gmail.com> - 2012-03-04 22:26 -0500
Re: Hardware for linked lists Andrew Reilly <areilly---@bigpond.net.au> - 2012-03-05 03:59 +0000
Re: Hardware for linked lists Anne & Lynn Wheeler <lynn@garlic.com> - 2012-03-04 23:38 -0500
Re: Hardware for linked lists Jim Haynes <jhaynes@alumni.uark.edu> - 2012-03-05 11:08 -0600
Re: Hardware for linked lists jgk@panix.com (Joe keane) - 2012-03-05 22:21 +0000
Re: Hardware for linked lists Michael S <already5chosen@yahoo.com> - 2012-03-05 15:14 -0800
Re: Hardware for linked lists John Levine <johnl@iecc.com> - 2012-03-06 01:44 +0000
Re: Hardware for linked lists Michael S <already5chosen@yahoo.com> - 2012-03-06 00:41 -0800
Re: Hardware for linked lists torbenm@diku.dk (Torben Ægidius Mogensen) - 2012-03-06 11:19 +0100
Re: Hardware for linked lists nmm1@cam.ac.uk - 2012-03-06 11:07 +0000
csiph-web