[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Living with the Chaos protocol

	Date: Fri, 3 Apr 92 11:18:46 EST
	From: delaney@xn.ll.mit.edu
	Message-Id: <9204031618.AA29300@pallas.ll.mit.edu>
	To: wlim@gdstech.grumman.com
	Cc: slug@ai.sri.com
	In-Reply-To: Willie Lim's message of Fri,
		     3 Apr 92 10:57:27 EST <9204031557.AA03870@gdstech.grumman.com>
	Subject: Living with the Chaos protocol
	Reply-To: delaney@xn.ll.mit.edu
	Having half-answered your question, I have to ask why your corporate
	people are against CHOAS traffic? the questions is partially


  Well, when you have thirty or fourty protocols running on the same link, you
can get some really weird interactions. Now, we all know that every vendor
writes well tested and well thought out code that follows all of the
applicable standards, so this shouldn't be a problem but for some reason
things just keep breaking. We've had problems like adding one more gator box
breaks all of the appletalk networks. Another favorite is machines that aren't
gateways repeating another protocols broadcast packets. Throughput does suffer
a bit when this happens. (The only way we've found to stop a broadcast storm
is to take all of the bridges down.) I've been arguing for years that we
should limit the backbone to only three or four protocols but in a
decentralized environment, one isn't allowed to tell others that they have to
spend money to translate protocols if they want to use the net. 

  Mike McDonald				Advanced Technology Dept.	
					Harris Corp.
  Email: mac@trantor.harris-atd.com	M.S. 3A-1912
  Voice: (407) 727-5060			P.O. Box 37
  Fax:   (407) 729-3363			Melbourne, Florida 32902