Article View: comp.hypercube
Article #33Re: Question on hypercube routing (channel usage)
From: Wen-King
Date: Tue, 14 Jul 1987 12:24
Date: Tue, 14 Jul 1987 12:24
29 lines
1829 bytes
1829 bytes
In article <278@hubcap.UUCP> pase%oregon-grad.csnet@RELAY.CS.NET (Douglas M. Pase) writes:
> 000 -> 001
< 000 -> 010
> 000 -> 001 -> 011
... etc
<The channel between nodes 0 and 1 is used in four different cases, between 0
>and 2 in two cases, and 0 and 4 in only one case. This exponential piling
<of messages into the lower numbered channels (which is compounded if one
>considers similar usage of channels by other nodes as well) means that in
<heavy message traffic there will be a lot of contention for the lower numbered
>channels where the channels corresponding to the higher bits go unused.
000 -(0)-> 001
000 -(1)-> 010
000 -(0)-> 001 -(1)-> 011
000 -(2)-> 100
000 -(0)-> 001 -(2)-> 101
000 -(1)-> 010 -(2)-> 110
000 -(0)-> 001 -(1)-> 011 -(2)-> 111
On the countrary, I counted that channels 0, 1, and 2 are each used
exactly 4 times, though they are not all channels coming out of node
0. The composite usage, assuming all nodes in the cube are doing
similar things, will be the same for all of the individual channels.
+--------------------------------------------------------------------------+
| Wen-King Su wen-king@vlsi.caltech.edu Caltech Corp of Cosmic Engineers |
+--------------------------------------------------------------------------+
Technical Details
Message-ID:
<290@hubcap.UUCP>Path:
arc1-ca.pugleaf.net!utzoo!utgpu!water!watmath!clyde!cbosgd!ucbvax!hplabs!gatech!hubcap!Wen-KingShow Full Headers
Complete Article Headers
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: arc1-ca.pugleaf.net!utzoo!utgpu!water!watmath!clyde!cbosgd!ucbvax!hplabs!gatech!hubcap!Wen-King From: Wen-King@hubcap.UUCP Newsgroups: comp.hypercube Subject: Re: Question on hypercube routing (channel usage) Message-ID: <290@hubcap.UUCP> Date: Tue, 14-Jul-87 08:24:03 EDT Article-I.D.: hubcap.290 Posted: Tue Jul 14 08:24:03 1987 Date-Received: Fri, 17-Jul-87 04:25:27 EDT Sender: fpst@hubcap.UUCP Lines: 29 Approved: hypercube@hubcap.clemson.edu