From neat.cs.toronto.edu!yonge.cs.toronto.edu!cs.toronto.edu!blam-owner Fri Jul 29 13:54:23 1994 Received: from neat.cs.toronto.edu ([128.100.3.2]) by pobox.cs.toronto.edu with SMTP id <166150>; Fri, 29 Jul 1994 13:54:19 -0400 Received: from yonge.cs.toronto.edu ([128.100.1.8]) by neat.cs.toronto.edu with SMTP id <48153>; Fri, 29 Jul 1994 13:54:13 -0400 Received: from bay.csri.toronto.edu ([128.100.1.1]) by yonge.cs.toronto.edu with SMTP id <62519>; Fri, 29 Jul 1994 13:53:22 -0400 Received: from ferret.csri.toronto.edu by bay.csri.toronto.edu id <4809>; Fri, 29 Jul 1994 13:53:07 -0400 From: Mart Molle To: blam@cs.toronto.edu Subject: Email reflector set up at Toronto for MAC layer study group Cc: mart@cs.toronto.edu Message-Id: <94Jul29.135307edt.4809@bay.csri.toronto.edu> Date: Fri, 29 Jul 1994 13:53:00 -0400 Status: R Hello everyone, This email reflector has been set up at the university of Toronto to support discussions about updating to the Ethernet MAC layer. This is the first item to be posted to the discussion. To post an item, simply mail it to "blam@cs.toronto.edu" and it will be delivered to everyone on the list. The initial list of names is the one I gathered on Thursday morning at the Orlando 802.3 plenary meeting where we agreed to set up a study group. To add or delete names, please send email to me (mart@cs.toronto.edu) and I will take care of it. In addition to the above administrivia, I have a couple of technical items. I. Bug fix in BLAM Technical Report ----------------------------------- Ron Crane discovered a slight error in my calculation of the time required under the standard Ethernet binary exponential backoff algorithm for a host to make 16 unsuccessful transmission attempts in a row for the same packet. (I had accidentally included a 16th backoff delay (of maximum length) at the very end, which I shouldn't have, which made the calculated value of the time come out slightly larger than it should be.) I have now fixed the bug, which affects only the following 3 pages of the BLAM Tech Report that I distributed at Orlando: Page 19 (The right-hand end of the P[n] curve [i.e., the "damped spring" part rather than the "diagonal sine wave" part] in Figure 6 rises slightly.) Page 20 (The calculation in the footnote is corrected.) Page 21 (The constant 256 to changed to 225 in the stability discussion.) The online copy of the T-R has been updated to include these changes. For those of you who want to keep your copies up to date without sacrificing trees unnecessarily, I have made a separate 3-page postscript file called TR.pages.ps that contains only those changed pages. You can pick it up via anonymous ftp from the host ftp.cs.toronto.edu in the directory "pub/blam". II. Smurph simulator available ------------------------------ In Orlando, Paul Nikolich asked me if I could make available the Ethernet simulator that I used to produce the experiments listed in the technical report. The necessary files have been gathered together in the same anonymous ftp directory ("pub/blam"). They include a Unix "tar" archive of version 1.12 of the Smurph simulation package produced at the University of Alberta, and the specific files we wrote to create a detailed Ethernet simulation model. Check the "README" file for more information. That's it for now. Thanks for you attention. Regards, Mart. ---------------------------------------------------------------------- Mart L. Molle, Professor Computer Systems Research Institute University of Toronto Toronto, Canada M5S 1A1 (416)978-4928 FAX: (416)978-4765 ----------------------------------------------------------------------