From braden at ISI.EDU  Mon May  3 15:54:46 2010
From: braden at ISI.EDU (Bob Braden)
Date: Mon, 03 May 2010 15:54:46 -0700
Subject: [ih] Fwd: Re: internet-reporters Majordomo Mailing List
Message-ID: <4BDF5436.4070707@isi.edu>



Our operations people just asked me about a very old ISI mailing list, 
internet-reporters at isi.edu.  I must admit I don't recall it, but I 
thought that other old timers would enjoy this list of old timers.  I 
wonder whether any of those addresses are still good?

Enjoy!

Bob Braden

_______________________________________

internet-reporters at isi.edu


Braden at isI.EDU,
Brescia at BBN.COM,
brian at hursley.ibm.com,
catlett at NCSA.UIUC.EDU,
CLynn at BBN.COM,
dcmwood at spot.colorado.edu,
ddc at LCS.MIT.EDU,
dfazio at MR.NET,
dfk at ripe.net,
dorl at VMS.MACC.WISC.EDU,
elliott at isi.edu,
ema at merit.edu,
epg at home.net,
eveleth at nwnet.net,
farrell at RICE.EDU,
fedor at PSI.COM,
nic at prep.net,
gary at airsoft.com,
guru at arl.wustl.edu,
hank at vm.tau.ac.il,
hankins at CIC.NET,
hastings at morgul.psc.edu,
jcurran at nic.near.net,
Jill.Foster at NEWCASTLE.AC.UK,
jon at CS.UCL.AC.UK,
jseeger at BBN.COM,
Kent at BBN.COM,
kirstein at cs.ucl.ac.uk,
kjk at spot.colorado.edu,
klh at nisc.sri.com,
Kseo at BBN.COM,
lefebvre at BBN.COM,
meeting-planning at ietf.org,
morris at WINDOM.UCAR.EDU,
pburns at yuma.acns.colostate.edu,
Postel at ISI.EDU,
R.Day at JNT.AC.UK,
raresec at rare.nl,
schoff at PSI.COM,
scoya at ietf.org,
SEstrada at aldea.com,
srh at merit.edu,
tjs at MSC.EDU,
tom at cic.net,
VCerf at mci.net,
wls at psi.com


From jmamodio at gmail.com  Tue May  4 05:48:24 2010
From: jmamodio at gmail.com (Jorge Amodio)
Date: Tue, 4 May 2010 07:48:24 -0500
Subject: [ih] Fwd: Re: internet-reporters Majordomo Mailing List
In-Reply-To: <4BDF5436.4070707@isi.edu>
References: <4BDF5436.4070707@isi.edu>
Message-ID: <u2k202705b1005040548vf58ccb2bl88b47fc51f60760b@mail.gmail.com>

On Mon, May 3, 2010 at 5:54 PM, Bob Braden <braden at isi.edu> wrote:
> ?I wonder whether any of those addresses are still good?

I'm sure that these two are not:

> eveleth at nwnet.net,
> nic at prep.net,

I was with VERIO when we acquired several of the old NSFNet regionals
after they transitioned to private/commercial, VERIO took
control/ownership of almost all associated domain names, some
addresses where still valid for a while or forwarded to various new
domains created by VERIO.

My .02
Jorge



From craig at aland.bbn.com  Fri May  7 10:20:41 2010
From: craig at aland.bbn.com (Craig Partridge)
Date: Fri, 07 May 2010 13:20:41 -0400
Subject: [ih] invention of multicast addressing
Message-ID: <20100507172041.5694628E137@aland.bbn.com>


Hi folks:

I'm trying to nail down when the concept of a "multicast address" came about
(just a little citation in a larger paper I'm writing).

In 1976 in the original Ethernet paper, there was unicast and broadcast.

In 1978, in their survey of local networks, Clark, Pogran and Reed mention
in passing that Mockapetris is playing around with bit wildcarding that
permits multiple addressees using one address.

In the 1980 Ethernet specification there are Ethernet multicast addresses as we 
know them today.

Digging a bit deeper from references in later papers, it appears that
Mockapetris, Lyle and Farber may have proposed a form of multicasting in 1977
(IFIP Congress paper of August 1977 that I don't have).

That suggests that someone saw the Mockapetris-Lyle-Farber idea, simplified it
and put it into the 1980 Ethernet standard (where it sat unused for several
years...).  But I can find no trail... Anyone got insights?

Thanks!

Craig


From jeanjour at comcast.net  Fri May  7 10:54:54 2010
From: jeanjour at comcast.net (John Day)
Date: Fri, 7 May 2010 13:54:54 -0400
Subject: [ih] invention of multicast addressing
In-Reply-To: <20100507172041.5694628E137@aland.bbn.com>
References: <20100507172041.5694628E137@aland.bbn.com>
Message-ID: <a06240822c80a031fd2b3@[10.0.1.15]>

What about Yogen Dalal's PhD thesis?  The first (that I know of) 
distributed spanning tree algorithms for multicast?  When I get home, 
I will check to see but I may be able to point to some mid- to late- 
70s discussions of it.  Although the title of Yogen's thesis says 
broadcast, I remember it as multicast, i.e. less than all!

But you may be on to something with the Irvine ring?

At 13:20 -0400 2010/05/07, Craig Partridge wrote:
>Hi folks:
>
>I'm trying to nail down when the concept of a "multicast address" came about
>(just a little citation in a larger paper I'm writing).
>
>In 1976 in the original Ethernet paper, there was unicast and broadcast.
>
>In 1978, in their survey of local networks, Clark, Pogran and Reed mention
>in passing that Mockapetris is playing around with bit wildcarding that
>permits multiple addressees using one address.
>
>In the 1980 Ethernet specification there are Ethernet multicast 
>addresses as we
>know them today.
>
>Digging a bit deeper from references in later papers, it appears that
>Mockapetris, Lyle and Farber may have proposed a form of multicasting in 1977
>(IFIP Congress paper of August 1977 that I don't have).
>
>That suggests that someone saw the Mockapetris-Lyle-Farber idea, simplified it
>and put it into the 1980 Ethernet standard (where it sat unused for several
>years...).  But I can find no trail... Anyone got insights?
>
>Thanks!
>
>Craig



From jack at 3kitty.org  Fri May  7 11:26:44 2010
From: jack at 3kitty.org (Jack Haverty)
Date: Fri, 07 May 2010 11:26:44 -0700
Subject: [ih] invention of multicast addressing
In-Reply-To: <20100507172041.5694628E137@aland.bbn.com>
References: <20100507172041.5694628E137@aland.bbn.com>
Message-ID: <1273256804.3524.41.camel@localhost>

Hi Craig,

>From some IBM-land document I found:

"...
Address---Contains the SDLC address of the secondary station, which
indicates whether the frame comes from the primary or secondary. This
address can contain a specific address, a group address, or a broadcast
address. A primary is either a communication source or a destination,
which eliminates the need to include the address of the primary.
..."

IIRC, IBM's SDLC was developed in the early/mid 70s.  Depending on
whether or not you think an SDLC "group address" fits the concept of
"multicast address", SDLC might be when the concept came about.  There
were other computer communications architectures at the time which
supported similar multiple-destination addressing, so IBM SDLC may not
have been the first.  Multi-point and multi-drop architectures were in
common use at the time because leased lines were so expensive.

I think you may find that a lot of architectural concepts like multicast
addressing have roots in the earlier vendor-proprietary computer
communications technologies, which have now mostly been steamrollered
into oblivion by The Internet.  It's very hard to tell which of all that
technology influenced the later work.

/Jack


On Fri, 2010-05-07 at 13:20 -0400, Craig Partridge wrote:
> Hi folks:
> 
> I'm trying to nail down when the concept of a "multicast address" came about
> (just a little citation in a larger paper I'm writing).
> 
> In 1976 in the original Ethernet paper, there was unicast and broadcast.
> 
> In 1978, in their survey of local networks, Clark, Pogran and Reed mention
> in passing that Mockapetris is playing around with bit wildcarding that
> permits multiple addressees using one address.
> 
> In the 1980 Ethernet specification there are Ethernet multicast addresses as we 
> know them today.
> 
> Digging a bit deeper from references in later papers, it appears that
> Mockapetris, Lyle and Farber may have proposed a form of multicasting in 1977
> (IFIP Congress paper of August 1977 that I don't have).
> 
> That suggests that someone saw the Mockapetris-Lyle-Farber idea, simplified it
> and put it into the 1980 Ethernet standard (where it sat unused for several
> years...).  But I can find no trail... Anyone got insights?
> 
> Thanks!
> 
> Craig



From LarrySheldon at cox.net  Fri May  7 11:34:05 2010
From: LarrySheldon at cox.net (Larry Sheldon)
Date: Fri, 07 May 2010 13:34:05 -0500
Subject: [ih] invention of multicast addressing
In-Reply-To: <20100507172041.5694628E137@aland.bbn.com>
References: <20100507172041.5694628E137@aland.bbn.com>
Message-ID: <4BE45D1D.8070500@cox.net>

On 5/7/2010 12:20, Craig Partridge wrote:
> Hi folks:
> 
> I'm trying to nail down when the concept of a "multicast address" came about
> (just a little citation in a larger paper I'm writing).
> 
> In 1976 in the original Ethernet paper, there was unicast and broadcast.

Well, I learned something.  I didn't know multicasting ("Mutlicasting
one of the cisco books calls it) was a a layer two protocol--all I've
ever read or heard about was at layer three--specifically at IP.

> 
> In 1978, in their survey of local networks, Clark, Pogran and Reed mention
> in passing that Mockapetris is playing around with bit wildcarding that
> permits multiple addressees using one address.
> 
> In the 1980 Ethernet specification there are Ethernet multicast addresses as we 
> know them today.
> 
> Digging a bit deeper from references in later papers, it appears that
> Mockapetris, Lyle and Farber may have proposed a form of multicasting in 1977
> (IFIP Congress paper of August 1977 that I don't have).
> 
> That suggests that someone saw the Mockapetris-Lyle-Farber idea, simplified it
> and put it into the 1980 Ethernet standard (where it sat unused for several
> years...).  But I can find no trail... Anyone got insights?
> 
> Thanks!
> 
> Craig
> 


-- 
Somebody should have said:
A democracy is two wolves and a lamb voting on what to have for dinner.

Freedom under a constitutional republic is a well armed lamb contesting
the vote.

Requiescas in pace o email
Ex turpi causa non oritur actio
Eppure si rinfresca

ICBM Targeting Information:  http://tinyurl.com/4sqczs
http://tinyurl.com/7tp8ml

	


From dave at farber.net  Fri May  7 11:36:00 2010
From: dave at farber.net (Dave Farber)
Date: Fri, 7 May 2010 14:36:00 -0400
Subject: [ih] invention of multicast addressing
In-Reply-To: <a06240822c80a031fd2b3@[10.0.1.15]>
References: <20100507172041.5694628E137@aland.bbn.com>
	<a06240822c80a031fd2b3@[10.0.1.15]>
Message-ID: <709E0D5A-109E-4E52-8B1A-AD0B2378D91F@me.com>

A critical part of the Irvine ring was a multiaddress capability. I  
will hunt the paper when I get home. Date about 1972



On May 7, 2010, at 1:54 PM, John Day <jeanjour at comcast.net> wrote:

> What about Yogen Dalal's PhD thesis?  The first (that I know of)  
> distributed spanning tree algorithms for multicast?  When I get  
> home, I will check to see but I may be able to point to some mid- to  
> late- 70s discussions of it.  Although the title of Yogen's thesis  
> says broadcast, I remember it as multicast, i.e. less than all!
>
> But you may be on to something with the Irvine ring?
>
> At 13:20 -0400 2010/05/07, Craig Partridge wrote:
>> Hi folks:
>>
>> I'm trying to nail down when the concept of a "multicast address"  
>> came about
>> (just a little citation in a larger paper I'm writing).
>>
>> In 1976 in the original Ethernet paper, there was unicast and  
>> broadcast.
>>
>> In 1978, in their survey of local networks, Clark, Pogran and Reed  
>> mention
>> in passing that Mockapetris is playing around with bit wildcarding  
>> that
>> permits multiple addressees using one address.
>>
>> In the 1980 Ethernet specification there are Ethernet multicast  
>> addresses as we
>> know them today.
>>
>> Digging a bit deeper from references in later papers, it appears that
>> Mockapetris, Lyle and Farber may have proposed a form of  
>> multicasting in 1977
>> (IFIP Congress paper of August 1977 that I don't have).
>>
>> That suggests that someone saw the Mockapetris-Lyle-Farber idea,  
>> simplified it
>> and put it into the 1980 Ethernet standard (where it sat unused for  
>> several
>> years...).  But I can find no trail... Anyone got insights?
>>
>> Thanks!
>>
>> Craig
>
>


From LarrySheldon at cox.net  Fri May  7 12:00:16 2010
From: LarrySheldon at cox.net (Larry Sheldon)
Date: Fri, 07 May 2010 14:00:16 -0500
Subject: [ih] invention of multicast addressing
In-Reply-To: <4BE45D1D.8070500@cox.net>
References: <20100507172041.5694628E137@aland.bbn.com>
	<4BE45D1D.8070500@cox.net>
Message-ID: <4BE46340.5000807@cox.net>

On 5/7/2010 13:34, Larry Sheldon wrote:

> Well, I learned something.  I didn't know multicasting ("Mutlicasting
> one of the cisco books calls it) was a a layer two protocol--all I've
> ever read or heard about was at layer three--specifically at IP.

Forgot it now seems more likely.

-- 
Somebody should have said:
A democracy is two wolves and a lamb voting on what to have for dinner.

Freedom under a constitutional republic is a well armed lamb contesting
the vote.

Requiescas in pace o email
Ex turpi causa non oritur actio
Eppure si rinfresca

ICBM Targeting Information:  http://tinyurl.com/4sqczs
http://tinyurl.com/7tp8ml

	


From dcrocker at gmail.com  Fri May  7 11:21:27 2010
From: dcrocker at gmail.com (Dave Crocker)
Date: Fri, 07 May 2010 11:21:27 -0700
Subject: [ih] invention of multicast addressing
In-Reply-To: <20100507172041.5694628E137@aland.bbn.com>
References: <20100507172041.5694628E137@aland.bbn.com>
Message-ID: <4BE45A27.7040300@gmail.com>

Hmmm.  I have a vague recollection of hearing that the Irvine Ring had some sort 
of multicast-ish capability.  Its addressing was based on process table lookup, 
rather than physical interface labels, so this seems a plausible memory.  I note 
that you copied Dave.

That, of course, would pre-date the Arpanet/Internet exploration of multicasting 
by a few years.

d/

On 5/7/2010 10:20 AM, Craig Partridge wrote:
> Hi folks:
>
> I'm trying to nail down when the concept of a "multicast address" came about
> (just a little citation in a larger paper I'm writing).
>
> In 1976 in the original Ethernet paper, there was unicast and broadcast.
>
> In 1978, in their survey of local networks, Clark, Pogran and Reed mention
> in passing that Mockapetris is playing around with bit wildcarding that
> permits multiple addressees using one address.
>
> In the 1980 Ethernet specification there are Ethernet multicast addresses as we
> know them today.
>
> Digging a bit deeper from references in later papers, it appears that
> Mockapetris, Lyle and Farber may have proposed a form of multicasting in 1977
> (IFIP Congress paper of August 1977 that I don't have).
>
> That suggests that someone saw the Mockapetris-Lyle-Farber idea, simplified it
> and put it into the 1980 Ethernet standard (where it sat unused for several
> years...).  But I can find no trail... Anyone got insights?
>
> Thanks!
>
> Craig
>

-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net


From p.schow at comcast.net  Fri May  7 12:20:27 2010
From: p.schow at comcast.net (Peter Schow)
Date: Fri, 7 May 2010 13:20:27 -0600
Subject: [ih] invention of multicast addressing
In-Reply-To: <a06240822c80a031fd2b3@[10.0.1.15]>
References: <20100507172041.5694628E137@aland.bbn.com>
	<a06240822c80a031fd2b3@[10.0.1.15]>
Message-ID: <20100507192026.GA2768@panacea.comcast.net>

On Fri, May 07, 2010 at 01:54:54PM -0400, John Day wrote:
> What about Yogen Dalal's PhD thesis?  The first (that I know of)
> distributed spanning tree algorithms for multicast?  When I get
> home, I will check to see but I may be able to point to some mid- to
> late- 70s discussions of it.  Although the title of Yogen's thesis
> says broadcast, I remember it as multicast, i.e. less than all!

Or the shorter more-accessible Dalal and Metcalfe CACM paper may 
provide some clues:

   Reverse path forwarding of broadcast packets
   Yogen K. Dalal (Xerox) and Robert Metcalfe (Stanford)
   Volume 21, Issue 12 (December 1978)

 "Broadcast multipoint communication is the delivery of messages to 
  all destinations, while multicast, or multidestination delivery, is 
  the delivery of messages to some specified subset of all the
  destinations. Ethernet, the DCS Ring and other broadcast networks are 
  ideally suited for such communication."

Vint Cerf was one of Dalal's Ph.D. supervisors, so I'll sure he'll
chime in soon.



From jnc at mercury.lcs.mit.edu  Fri May  7 18:25:22 2010
From: jnc at mercury.lcs.mit.edu (Noel Chiappa)
Date: Fri,  7 May 2010 21:25:22 -0400 (EDT)
Subject: [ih] invention of multicast addressing
Message-ID: <20100508012522.C0C1E6BE5C2@mercury.lcs.mit.edu>

    > From: Craig Partridge <craig at aland.bbn.com>

    > I'm trying to nail down when the concept of a "multicast address" came
    > about

Wasn't there some ARPANET work which had some concept of group addresses? I
think it was called "logical addressing"? As I recall it was defined, but
never implemented? Does anyone else have anything on this, or do I need to go
search? On a quick search online, I found this:

  "ARPANET Routing Algorithm Improvements"
  Semiannual Technical Rept. No. 2, 18 Apr-15 Oct 78,
  McQuillan, J. M.; Richer, I.; Rosen, E. C.; Bertsekas, D. P.
  BBN, Oct 1978

which includes this:

  A preliminary design of enhanced message capabilities for the ARPANET
  (logical addressing, broadcast addressing, and group addressing) is
  presented.

"Group addressing" sounds like what we now call multicast, no? Anyway, that
would be my bet for one of the earliest (other than proprietary things like
the IBM thing that Jack turned up).


    > In 1978, in their survey of local networks, Clark, Pogran and Reed
    > mention in passing that Mockapetris is playing around with bit
    > wildcarding that permits multiple addressees using one address.
    > ...
    > ... it appears that Mockapetris, Lyle and Farber may have proposed a
    > form of multicasting in 1977 (IFIP Congress paper of August 1977 that I
    > don't have).

The UC Irvine stuff wasn't really multi-cast as we now think of it. The
hardware implementation is described at a high level in:

  IEN-82, "LCS Net Address Format"

since the MIT V1 Ring was the Irvine ring (MIT actually got that version of
the ring to work, from a desk design that Paul Mockapetris did).

The way it worked was that packets had a 32-bit destination address and a
32-bit destination address mask. Each interface had a loadable table of 8 name
entries, each of which also had the paired 32-bit fields. For each bit in a
packet's destination address, for each of the 8 name entries: if either of the
two mask bits (i.e. the one in the packet and the one in the name entry) was
on, that bit 'matched'; if neither was on, but the two address bits matched,
that bit 'matched'; if all 32 bits of at least one name entry 'matched', the
entire destination address in the packet matched, and the packet was accepted
(i.e. received).

Obviously, that's very general-purpose, and we could easily have done
something with the semantics of multicasting (as we now think of it) with it.
I think it was originally designed to allow proceses to move invisibly among
hosts on the ring at Irvine, IIRC.

As you can see from the document, we thought of having the interfaces in
routers detect packets destined for off-local-link destinations and
automagically suck them up and forward them - sort of like a bridge. We never
actually implemented that, IIRC. (If anyone cares why, ask...) In general, we
never found a use for the powerful addressing stuff, and it was dropped in
the MIT V2 ring in favour of a simple 8-bit host number.

	Noel


From jnc at mercury.lcs.mit.edu  Fri May  7 18:41:02 2010
From: jnc at mercury.lcs.mit.edu (Noel Chiappa)
Date: Fri,  7 May 2010 21:41:02 -0400 (EDT)
Subject: [ih] invention of multicast addressing
Message-ID: <20100508014102.6D4B06BE5C8@mercury.lcs.mit.edu>

    > From: Dave Farber <dave at farber.net>

    > A critical part of the Irvine ring was a multiaddress capability. I
    > will hunt the paper when I get home. Date about 1972

I think I must have forgotten some earlier UC Irvine work when I made this
comment:

    >> ... it appears that Mockapetris, Lyle and Farber may have proposed a
    >> form of multicasting in 1977 (IFIP Congress paper of August 1977 that I
    >> don't have).

    > The UC Irvine stuff ... The hardware implementation is described at a
    > high level in:
    >  IEN-82, "LCS Net Address Format"
    > since the MIT V1 Ring was the Irvine ring

since it sounds from the above that there was an earlier UC Irvine ring (in
the early 70s), and the Mockapetric design that MIT implemented (starting in
the fall of '77) was a second-generation design?

	Noel


From esuarez at fcaglp.fcaglp.unlp.edu.ar  Fri May  7 20:39:45 2010
From: esuarez at fcaglp.fcaglp.unlp.edu.ar (Eduardo A. =?iso-8859-1?b?U3XhcmV6?=)
Date: Sat, 08 May 2010 00:39:45 -0300
Subject: [ih] invention of multicast addressing
In-Reply-To: <20100507192026.GA2768@panacea.comcast.net>
References: <20100507172041.5694628E137@aland.bbn.com>
	<a06240822c80a031fd2b3@[10.0.1.15]>
	<20100507192026.GA2768@panacea.comcast.net>
Message-ID: <20100508003945.4whw0l1jc4kco8o4@fcaglp.fcaglp.unlp.edu.ar>

Quoting Peter Schow <p.schow at comcast.net>:

> Or the shorter more-accessible Dalal and Metcalfe CACM paper may
> provide some clues:

The paper on CACM by Dalal refers to another paper by Paoletti on AUTODIN

Paoletti, L.M. AUTODIN. In Computer Communication Networks, R.L.  
Grimsdale and F.F. Kuo, Eds. (Proc. NATO Advanced Study Inst. Comptr.  
Comm. Networks, Sussex, U.K., Sept. 1973), Noordoff Int. Publ.,  
Leyden, 1975.

"Each ASC has the capability of accepting traffic from, at most, 200  
subscriber terminals which can vary in speed from 45 to 4800 bits per  
second. The average message length is 2,000 characters; the maximum  
message length is 40,000 characters. The average address multiplicity  
per message is 1.75."

The URL of that paper is http://www.rogerdmoore.ca/PS/ADINC.html

Eduardo.-



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.



From dcrocker at gmail.com  Sat May  8 09:57:32 2010
From: dcrocker at gmail.com (Dave Crocker)
Date: Sat, 08 May 2010 09:57:32 -0700
Subject: [ih] invention of multicast addressing
In-Reply-To: <20100508014102.6D4B06BE5C8@mercury.lcs.mit.edu>
References: <20100508014102.6D4B06BE5C8@mercury.lcs.mit.edu>
Message-ID: <4BE597FC.80808@gmail.com>

b

On 5/7/2010 6:41 PM, Noel Chiappa wrote:
>      >  From: Dave Farber<dave at farber.net>
>
>      >  A critical part of the Irvine ring was a multiaddress capability. I
>      >  will hunt the paper when I get home. Date about 1972
>
> I think I must have forgotten some earlier UC Irvine work when I made this
> comment:
>
>      >>  ... it appears that Mockapetris, Lyle and Farber may have proposed a
>      >>  form of multicasting in 1977 (IFIP Congress paper of August 1977 that I
>      >>  don't have).
>
>      >  The UC Irvine stuff ... The hardware implementation is described at a
>      >  high level in:
>      >   IEN-82, "LCS Net Address Format"
>      >  since the MIT V1 Ring was the Irvine ring
>
> since it sounds from the above that there was an earlier UC Irvine ring (in
> the early 70s), and the Mockapetric design that MIT implemented (starting in
> the fall of '77) was a second-generation design?


As I recall, the MIT version of the Irvine ring tossed out the process 
addressing table, which had taken about a quarter of the board's real estate. 
This would have eliminated the very natural "multiaddress" potential of the 
ring, although it could have been present in a different form of course.

I've copied Ken Porgran, in case he can offer some first-hand clarity.

d/

-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net


From dave at farber.net  Sat May  8 12:01:43 2010
From: dave at farber.net (David Farber)
Date: Sat, 8 May 2010 15:01:43 -0400
Subject: [ih] invention of multicast addressing
In-Reply-To: <4BE597FC.80808@gmail.com>
References: <20100508014102.6D4B06BE5C8@mercury.lcs.mit.edu>
	<4BE597FC.80808@gmail.com>
Message-ID: <853B6455-9533-4305-87D1-728DDD72B340@farber.net>

Indeed the version of the Irvine Ring we created at UC Irvine had process oriented addressimg and multi-address capabilities since it was designed to be part of a  fully distributed computer system ( a cloud :-) ). MIT was more interested in its use as a local network and removed the process  oriented addressing.


Dave


From jnc at mercury.lcs.mit.edu  Sat May  8 14:24:07 2010
From: jnc at mercury.lcs.mit.edu (Noel Chiappa)
Date: Sat,  8 May 2010 17:24:07 -0400 (EDT)
Subject: [ih] invention of multicast addressing
Message-ID: <20100508212407.E6D926BE5F3@mercury.lcs.mit.edu>

    > From: Dave Crocker <dcrocker at gmail.com>

    >> there was an earlier UC Irvine ring (in the early 70s), and the
    >> Mockapetric design that MIT implemented (starting in the fall of '77)
    >> was a second-generation design?

    > As I recall, the MIT version of the Irvine ring tossed out the process
    > addressing table

Ah, no, that was the _second_ ring done at MIT (the so-called 'V2' ring, a
10MBit/second ring done by an MIT/Proteon collaboration), which deleted all
the loadable addressing stuff.

The first ring done at MIT (the so-called 'V1' ring, a 1MBit/second design)
_was_ an Irvine design, and _did_ have all the loadable addressing stuff.

My point (above) was that that 'first' ring at MIT was actually the second
Irvine-designed ring, and there had been an _earlier_ one, at Irvine only.
Exactly what _that_ ring had I don't know, although I have this vague memory
that it had some sort of loadable addressing stuff - but someone who is
familiar with it will have to provide exact details. The details in my first
message (about the 32-bit name and mask) are accurate for the so-called 'V1'
ring. 

The two 'V1' and 'V2' rings were very different in many other significant
ways too - e.g. the method of dealing with clock variations between stations
(both in speed and phase) were totally different.


    > which had taken about a quarter of the board's real estate.

A little more than 1/4, I think, but I don't recall exactly. Ken might
remember in detail.

(And the V1 ring wasn't a plug-in board, but an entire half-height rack unit,
although inside it was just one big horizontal wire-wrapped board. I think
only about 8 were ever made, of which a couple [two, I think] went to UCLA.)

	Noel


From jtk at depaul.edu  Mon May 24 11:27:59 2010
From: jtk at depaul.edu (John Kristoff)
Date: Mon, 24 May 2010 13:27:59 -0500
Subject: [ih] Why UDP/DNS limited to 512 exactly?
Message-ID: <20100524182759.GA1453@condor.depaul.edu>

I'm curious what the reason was for the specific limit of 512 bytes
for a UDP/DNS message.

Was it a convenient value that fell on a word boundary and close
enough to the 576 IP datagram limit?

I would note in theory you could end up with a 512 byte DNS message
that wouldn't fit into a 576 byte datagram (after adding 8 bytes for
UDP and 60 bytes for a maximum IPv4 header with a full set of options).

John


From jabley at hopcount.ca  Tue May 25 07:48:31 2010
From: jabley at hopcount.ca (Joe Abley)
Date: Tue, 25 May 2010 16:48:31 +0200
Subject: [ih] Why UDP/DNS limited to 512 exactly?
In-Reply-To: <20100524182759.GA1453@condor.depaul.edu>
References: <20100524182759.GA1453@condor.depaul.edu>
Message-ID: <0BB2A08D-82ED-4B71-B0B2-9433C2075969@hopcount.ca>


On 2010-05-24, at 20:27, John Kristoff wrote:

> I'm curious what the reason was for the specific limit of 512 bytes
> for a UDP/DNS message.
> 
> Was it a convenient value that fell on a word boundary and close
> enough to the 576 IP datagram limit?

By some freak of chance I looked this up earlier this week as part of a document I was scratching together to describe the interaction between fragmentation and DNS responses. I didn't come up with a good reference. I would also be interested in where that specific number came from.

However, in the absence of any better theory I had assumed it was just...

> I would note in theory you could end up with a 512 byte DNS message
> that wouldn't fit into a 576 byte datagram (after adding 8 bytes for
> UDP and 60 bytes for a maximum IPv4 header with a full set of options).

... a safe bet given that it's rare for any IPv4 datagram to carry a full set of options (or, really, any at all), 512 + 8 + 20 = 540.


Joe