🚀 go-pugleaf

In article <328@hubcap.UUCP> bradley@M.CS.UIUC.EDU writes:
>I wouldn't mind seeing that list of references if it wouldn't be too much 
>trouble.  Sounds interesting.
>
>David Bradley
>
>bradley@a.cs.uiuc.edu
>Department of Computer Science
>University of Illinois at Urbana-Champaign

Some references that I can contribute :

The problem of deadlock in routing messages in a network is
part of the more general problem - Flow Control

The objectives of flow control in a network are:

1) Prevent throughput & response degradation due to network overload
2) Avoid deadlock
3) Allocate resources fairly to competing users

Store & Forward deadlocks have been studied in great detail in computer
n/w's. One commonly accepted solution to the problem of indirect
store & forward deadlocks (the situation described in one of the previous
messages) is by the use of "structured buffer pools". 

Some review papers where one can find numerous references to the actual work are

"Flow Control : A Comparitive Survey" Mario Gerla & Leonard Kleinrock
IEEE Trans. on Computer Communications, April 1980. 

"Prevention of Deadlocks in Packet-Switched Data Transport Systems" 
Klaus D. Gunther
IEEE Trans. on Comp. Comm, April 1981. 
{This entire issue is devoted to congestion control}

A recent paper which discusses deadlocks when "wormhole" routing (similar to
Kleinrock's "virtual cut through") is used is

"Deadlock-Free Message Routing in Multiprocessor Interconnection Networks"
W. J. Dally & Charles L. Seitz
IEEE Trans. on Computers, May 1987.

ravi

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
T. M. Ravi 
UCLA Computer Science Department, 3680 Boelter Hall, UCLA
Los Angeles, CA 90024	          Phone: (213) 825-2266
ARPA : ravi@CS.UCLA.EDU
UUCP : {...sdcrdcf, ihnp4, trwspp, ucbvax}!ucla-cs!ravi
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~