[ih] QUIC story

[Craig Partridge]

Wasn't just router silicon -- it was router design.  What made ATM
appealing is that it made the inside of the router or switch parallel,
which was necessary to push into multigabit rates. Folks had to figure out
how to rework an Internet router to be parallel and it took at least two
major innovations: fully-standalone forwarding tables  with associating
forwarding engines and breaking packets apart (essentially into cells),
squirting those parts through the parallel backplane, and then reassembling
the packet at the outbound interface for transmission.  Arguably there were
third and fourth innovations: third innovation was data structures that
gave good lookup times in the forwarding tables (e.g. the WashU and Lulea
algorithms), and  fourth innovation was to figure out how to collect the
SNMP data efficiently (in hardware and software) across fully-distributed
forwarding engines (routing SNMP data, at speed, through the cards and
collecting it over the packet bus was painful -- Phil Carvey figured this
one out and just put a high speed Ethernet in the box exclusively for
SNMP).  Two teams, one at BBN and one at Juniper, solved the first and
second problems - independently coming to similar solutions.  Third
innovation was largely in academia motivated by some talks I gave at the
time (I was leading the BBN team - Tony Li led the Juniper team).  Last one
was Phil Carvey (lead hardware guy on the BBN team and then principle at
Avici).

Craig

On Sat, Jun 25, 2022 at 3:27 PM Andrew G. Malis via Internet-history <
internet-history at elists.isoc.org> wrote:

> Brian,
>
> Yeah, back then the joke was that McQuillan was the only one making money
> from ATM. :-) That did change in a big way (for a while) in the last 90s
> and early 2000s, before router silicon caught up.
>
> Cheers,
> Andy
>
>
> On Sat, Jun 25, 2022 at 4:55 PM Brian E Carpenter via Internet-history <
> internet-history at elists.isoc.org> wrote:
>
> > Three points:
> >
> > 1) It was way beyond an "academic experiment" by 1988/89, IMHO. When IBM
> > funded the first transatlantic T1 in 1990, it was already a production
> > network for the academic and research community. In fact, that's exactly
> > why the NSFnet AUP existed, and why IBM threw in substantial funding
> > for the T1.
> >
> > 2) When was the fuss about registering 3com.com? It wasn't so much the
> > issue of a domain starting with a digit, but the issue of a domain
> > being equal to a trademark that was controversial, I think. Anyway,
> > it was a sign of the times.
> >
> > 3) In the anecdote department, I recall taking a day off from my first
> > IETF meeting in 1992 (#25, in D.C.) to go across town to attend a one-day
> > McQuillan conference (on ATM??). The funny thing was that almost all the
> > speakers wearing suits were people I'd seen the day before at the IETF in
> > jeans and T-shirts.
> >
> > Regards
> >     Brian Carpenter
> >
> > On 26-Jun-22 03:38, Jorge Amodio via Internet-history wrote:
> > > Hi Dave,
> > >
> > > Agreed, that is my recollection as well when I got remotely involved in
> > the
> > > mid 80's/early 90's. There was in fact some aversion to having vendors
> > > participate in meetings, I believe on our side some of that sentiment
> was
> > > partially driven by NSF's AUP and that the Internet was mostly an
> > academic
> > > experiment.
> > >
> > > -J
> > >
> > >
> > > On Sat, Jun 25, 2022 at 7:30 AM Dave Crocker via Internet-history <
> > > internet-history at elists.isoc.org> wrote:
> > >
> > >> On 6/25/2022 5:11 AM, Jorge Amodio via Internet-history wrote:
> > >>> I'm not sure where you get your information from, but vendors have
> been
> > >>> deeply involved since the early days of the Internet, even ARPANet,
> > BBN,
> > >>> Cisco, DEC, etc, were private companies and "vendors" since their
> > >> inception
> > >>> and there has been a constant participation from companies and
> services
> > >>> providers for very long time.
> > >>
> > >>
> > >> There was something of a milestone, in this regard, around 1987. Prior
> > >> to that, vendor participation was from a strongly-linked relationship
> to
> > >> am Arpanet/Internet research contractors, or even from aDirect
> > >> government contract  Permission-by-association, if you will.
> > >>
> > >> After that, random commercial representatives were permitted to attend
> > >> IETF meetings.
> > >>
> > >> Not the sort of thing to add to a resume, but I turned out to be the
> > >> test case that produced this change.
> > >>
> > >> I was working for a company that produced after-market TCP/IP stacks.
> > >> We had no direct involvement in any Internet R&D. Just a company
> selling
> > >> its wares.  Given how rapidly Internet tech was changing at that
> time, I
> > >> wanted us attending IETF meetings.
> > >>
> > >> The IETF initially rejected the request, but I pressed.  Much
> discussion
> > >> ensured, and I believe the decisive comment was Bob Braden's that was
> > >> along the lines of "come on folks, it's Dave, and we know him."
> > >>
> > >> This was utterly irrelevant logic, but apparently swayed IETF folk
> > >> enough for permission to be granted.  So I got to attend.  By the
> > >> meeting after that, the floodgates were fully opened, with other
> vendors
> > >> attending.
> > >>
> > >> In spite of compelling reasons to motivate one, I remain steadfastly
> > >> unapologetic...
> > >>
> > >>
> > >> d/
> > >>
> > >> --
> > >> Dave Crocker
> > >> Brandenburg InternetWorking
> > >> bbiw.net
> > >>
> > >> --
> > >> Internet-history mailing list
> > >> Internet-history at elists.isoc.org
> > >> https://elists.isoc.org/mailman/listinfo/internet-history
> > >>
> > --
> > Internet-history mailing list
> > Internet-history at elists.isoc.org
> > https://elists.isoc.org/mailman/listinfo/internet-history
> >
> --
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
>


-- 
*****
Craig Partridge's email account for professional society activities and
mailing lists.