Talk:Hoard memory allocator

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
Low	This article has been rated as Low-importance on the project's importance scale.
	This article is supported by WikiProject Software (assessed as Low-importance).

This article looks like shameless self-promotion. Provide proofs of claim(s) or remove the page — Preceding unsigned comment added by 67.70.246.110 (talk) 2007-03-21T15:58:42

(Originally asked on Talk:Bisqwit:

Hi Bisqwit, could you please provide some more precision as to the 'weasel words' notation on the Hoard page, since it appears that all comments are based on the cited scientific article. Thanks. --83.44.209.167 (talk) 13:01, 11 March 2009 (UTC) Emery Berger[reply]

Quantitive expressions like "fast", "low fragmentation" and comparative expressions "improve" and "reduce" only make sense if there is a reference point. They are often used in advertising context because of the impression they create without anything concrete. You can make the text more neutral by using quotative expressions such as "aims for low fragmentation" and "claims to improve". A scientific article linked does not offset these problems. Also, the scientific article is apparently written by yourself, which is something on the verge of "original research" and not well-taken in general in Wikipedia. --Bisqwit (talk) 15:17, 12 March 2009 (UTC)[reply]

Actually, no. The paper in question is peer-reviewed, and published at one of the most prominent venues in computer science research. Hoard is heavily cited and many papers refer to it as state of the art. The reference points used in the original paper include a number of other memory allocators (including the one used by Linux) across a range of applications. The claims are supported both by evidence and by mathematical proofs. 147.83.181.28 (talk) 18:47, 12 March 2009 (UTC)Emery Berger[reply]

Article rewritten[edit]

Please review my new version of the article to fix inconsistencies, mistakes and other things.

I agree with Bisqwit, the previous version of this article was shameless self-promotion because based on an old scientific article. After 2000, other memory allocators have improved their behaviour on multiprocessor architecture due to incoming multicore PC (e.g. Core Duo). Wikipedia cannot say Hoard is the best memory allocator without peer-reviewed recent benchmarks.

--Oliver.hessling (talk) 13:26, 12 June 2009 (UTC)[reply]

fastest[edit]

I am currently looking for the "best" allocator to use in my application and stumbled over this, among other articles. Unfortunately, without looking into the cited papers, it is not possible to judge anything based on these articles. The papers have some numbers, why aren't those mentioned in the article? like "1.4 times faster with the famous threadtest than hoard on 8 cores, measured in 2002 on an xxx cpu". This would give people at least rough directions, or allows them to start researching/testing the appearant fastest allocator.

Another thing that I wonder about this is, the article claims that there is a new allocator better than hoard, but after that that the author worked on hoard to improve. Is hoard now as fast as that new one? or is the new one still faster, and where is it? —Preceding unsigned comment added by 213.61.9.75 (talk) 08:07, 17 May 2011 (UTC)[reply]