Talk:Fat tree

	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 Networking task force (assessed as Mid-importance).

Telecommunications Mid‑importance

	Telecommunication portal This article is within the scope of WikiProject Telecommunications, a collaborative effort to improve the coverage of Telecommunications 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.TelecommunicationsWikipedia:WikiProject TelecommunicationsTemplate:WikiProject TelecommunicationsTelecommunications articles
Mid	This article has been rated as Mid-importance on the project's importance scale.

Related topologies?[edit]

What purpose does the Related Topologies section serve? It reads like two editors bickering about semantics and contributes nothing to my understanding of what a fat tree is. The only thing I take away from it is that some paper published by some group used some wrong words. Is that topology used anywhere? Does anyone in production networking even care about that topology? It seems amateur and I think it should be removed. — Preceding unsigned comment added by 132.249.64.124 (talk) 16:33, 8 February 2013 (UTC)[reply]

A dash of trouble[edit]

Is it a fat tree, or a fat-tree? ~ Jafet (spam) 13:39, 19 August 2007 (UTC)[reply]

what is a fat tree and what is not[edit]

or ?? `a5b (talk) 04:05, 4 July 2010 (UTC)[reply]

acc to http://www.ics.forth.gr/~kateveni/534/03a/s52_benes.html -- both are. but the second is only one of implementation. `a5b (talk) 06:04, 4 July 2010 (UTC)[reply]

True. In fact, this would be a nice topic (plus illustration) to add to the article, the logical view of a fat tree versus its usual implementation. Scott Pakin (talk) 07:25, 16 December 2010 (UTC)[reply]

For better or worse, the hyphen in "fat-tree" has always been there.

I see two problems with this article: The claim that the UCSD authors made a "misleading" claim, and then the explanation that their network is not a fat-tree because instead of having single "fat" channels to parents, channels are implemented using parallel links. From the very first paper by Leiserson, the channels have been composed of multiple parallel links. (I.e., the "logical view" and the "implementation" have always existed together.) Also, various Clos networks may be isomorphic to fat-trees. — Preceding unsigned comment added by 152.3.136.127 (talk) 22:35, 14 January 2013 (UTC)[reply]

Bottlenecks and SPOF?[edit]

Doesn't the root node act as a single point of failure? Isn't that the case for all centralized networks like star? The central node of a star acts as a single point of failure, wouldn't the same apply for the root node of a fat-tree? Don't all centralized networks also have bottlenecks? Don't star and hierarchical tree networks have bottlenecks? Mesh, hypercube, and torus networks avoid bottlenecks. WorldQuestioneer (talk) 17:11, 19 July 2020 (UTC)[reply]