Re: "There's a routing problem I have glossed over..."
Gregory Alan Bolcer (gbolcer@gambetta.ICS.uci.edu)
Thu, 08 Oct 1998 12:57:51 -0700
> Shepard's thesis, under Dave Clark, considered the routing problem in rando=
> arrays of relays. His major finding was that routing hotspots were
> unavoidable, generally in the physical center of the network. Congestion
> would overcome.
I think a couple of worm-infested fat trees would attack the
problem nicely. It's a very elegant solution for supporting
fine-grained parallelism and message passing in hypercubes
of processors. Wow, maybe I did learn something in grad school.
Greg "Tree hugger and Worm Eater"
Ronald I. Greenberg and Charles E. Leiserson. Randomized Routing on
Fat-Trees. In IEEE 26th Annual Symposium on the Foundations of Computer
Science. IEEE, November 1985.
Ronald I. Greenberg and Lee Guan, An Improved Analytical Model for
Wormhole Routed Networks with Application to Butterfly Fat-Trees,