Re: [Coin] Comments re. draft-irtf-coinrg-use-cases (was: Re: Cancelling the COINRG meeting at IETF113)

[Marie-Jose Montpetit <[e-mail]>]

Toerless:

The work of the RG is not only about use cases. You are evaluating the work
of a RG like the one of a WG namely in the number of drafts (and BTW we
have a few others). Producing drafts is not the goal. We have a strong
recommendations from Colin to create an agora for researchers to come and
present their research. That may or may not result in drafts. We also had a
number of hackathons (and we may start them again now that P4 has been
ported to PIs) to try out some “in network” ideas in P4. I would refer you
to the work presented in our meetings in the past 3 years and please tell
me that they are only use cases on distributed applications.

As per the use case it is an attempt to see what happens to a number of
applications once that computing is available in the network elements. You
are not the first one to comment on that. There has been a lof discussion
about the difference between in network (in the router) and adding a side
device. Of course the side device becomes an edge and there is a fine line.
I would refer to yo the now seminal Netcache paper that combines PISA
switching and GPU. And Dave Oran also discussed on this list and in
meetings about the difference between “in” and “on” network. That’s the
beauty of a RG we can discuss this and do not have to produce drafts. If
you have specific recommendations for the draft please make them. As a
co-author I note your recommendations.

Personally my view for COIN is that of an Internet that resembles a
computer board (not a telephone network) and we are in someway defining its
elements and its operating system. Traditional operations including routing
may take a new life. And I tell my students to think of the “life of a
packet” when network elements can modify the packet not just forward them
(from the stack in their phones all the way to the data centers). Noa
Zilberman of Oxford U. talks of computing service providers when computing
is available in the network in RANs for example. There is a large community
of researchers looking at how to create new paradigms for assembling
compute/forward elements to offer new functionalities (in video for
example). In that context many concepts may change, including routing. This
is the research I am interested in. Maybe there will be drafts. Maybe not.

I note the suggestion for a name change but I do not think it will happen
based on your (mis)understanding of one draft.

And I would respectfully request a toning down of your snide comments. You
may not think our work is to your liking and this is fine. You may think we
need people like you to anchor of blue sky thoughts in reality and we thank
you but a more positive approach would yield better results. Criticism is
too easy.. And if you think the RG is not doing quality work or the the
chairs are incompetent or ineffectual raise it with Colin and even the the
IAB. We are ready for any review.



mjm
Marie-José Montpetit, Ph.D.
[e-mail]



From: Toerless Eckert <[e-mail]> <[e-mail]>
Reply: Toerless Eckert <[e-mail]> <[e-mail]>
Date: March 13, 2022 at 3:33:21 AM
To: Sharon Barkai <sharon.barkai=[e-mail]>
<sharon.barkai=[e-mail]>
Cc: Schooler, Eve M <[e-mail]>
<[e-mail]>, Marie-Jose
Montpetit <[e-mail]> <[e-mail]>, coinrg-chairs
<[e-mail]> <[e-mail]>, Dirk Trossen
<dirk.trossen=[e-mail]>
<dirk.trossen=[e-mail]>, coin <[e-mail]>
<[e-mail]>
Subject:  [Coin] Comments re. draft-irtf-coinrg-use-cases (was: Re:
Cancelling the COINRG meeting at IETF113)

Trying to bring back the reply to what Sharon observes to the RGs work,
which so
far officially only includes the use-cases it seems.

As a very high-level observation, i find it less than ideal to prefix
everything
in that document equally just with "COIN" because it eliminates important
differences
and attaches unnecessary a new term to something that already is well
known. It would
also be good to remember that just because an RG has a particular name, it
is not
necessary for all technical classiciations to re-use that RGs name.

To me, the mayority of use-cases presented is really "just" distributed
applications,
which in my view just "use" the network, but which are not "in" the
network. Aka:

These use-cases run predominantly on general-purpose compute
(x86/arm/risc5) and this
compute is somehow distributed and may include mobile components (like
user-endpoints).
And we called this distributed applications for decades without anyone ever
complaining
about that term.

These applications need some varying degree of better-than-best-effort
services from the
network, such as controlled or guaranteed throughput, latency, loss and
availability,
and they also may need some multipoint packet delivery, and some discovery
functions from
the network to seed their self-orchestration. But that set of requirements
does in my book
not make them "in" the network. That set of requirements existed for
decades as well.

Another example: if a vendor like Cisco or Huawei sells a side-edge-device
consisting of a
VM/container host system and you can separately instantiate a router, a
firewall, a DNS,
an email, a web and a bunch of other servers: That to me is not "compute in
the network".
That is just softwareization to combine decade old functions/devices into a
single box.

So, to me, 5.3, (Virtual Network Programming), is the only proper "in
network" case
described in the document.

Now, my understanding of what's in and whats not in the network might be
different
from what the RG mayority wants, but at least it would be great to spend
more time
with a somewhat longer list of examples and explain for each of them
whether why
and how its considered to be in or out, if in and out is really what the RG
wants to define.

IMHO, it would be more productive to come up with a more differentiated set
of classifications.
Distributed applications and their needs for better network services do not
become less
important by NOT giving them a new name COIN.

One could also simply rename the RG to "Computer Over and In the Network"
if one feels
the risk of kicking all the interesting work out of scope by not declaring
it to be "in".

Cheers
Toerless

On Sat, Mar 12, 2022 at 10:51:14PM +0200, Sharon Barkai wrote:
> There is some duality in the list between those focusing on making
switches/routers more like computers, and those focusing on using the
network cloud as a well .. a cloud - for when it fits - topology, sharding,
privacy etc.
>
> In my view these are simply bottom-up top-down sides of the same you
know… so im sure the chairs will settle this in time with proper
frameworks.
>
> We needn't start from scratch on neither. Theres been good existing
proposals for baseline on both fronts already.
>
> --szb
> Cell: +972.53.2470068
> WhatsApp: [phone]
>
> > On Mar 11, 2022, at 10:36, Schooler, Eve M <[e-mail]>
wrote:
> >
> > 
> > Hi Dirk, All,
> >
> > My apologies about the ambiguity in the comment about the agenda. It
was intended to convey that we struggled to have a FULL agenda, and NOT to
pass judgement on the quality of the topics that might have been presented.
Of the individuals we reached out to present, many stated the day/timing
simply did not work out.
> >
> > As for transparency…If you are a regular reader of this list, then it
is painfully obvious that there has been quite a bit of divisiveness
happening both on and off the list. As chairs, given the state of the
agenda and the tone of the dialog, we felt the need to take a step back
from the vitriol and simply take a deep breath to regroup.
> >
> > We certainly have valued the continued involvement of the COIN
community, which has made many of the discussions vibrant and rewarding.
> >
> > Best regards,
> > Eve
> >
> > From: Coin <[e-mail]> On Behalf Of Dirk Trossen
> > Sent: Thursday, March 10, 2022 10:30 PM
> > To: Marie-Jose Montpetit <[e-mail]>; coin <[e-mail]>
> > Cc: coinrg-chairs <[e-mail]>
> > Subject: Re: [Coin] Cancelling the COINRG meeting at IETF113
> >
> > Hi J/E/M, all,
> >
> > Now that’s a surprise, not just in content but also in style since the
RG community lacks the transparency of this decision.
> >
> > As a COIN RG member myself for now more than 3 years (spanning two
organizations), I had looked forward to discussing at least three
activities in which I am involved in, namely the (i) use case advances
(trying to formulate and categorize the pertinent research questions in a
number of COIN areas), (ii) the applicability of SDN for routing (i.e. the
use of DP programmability for realizing novel routing solutions, which
according to the chairs is in scope of COIN), and (iii) a discussion on how
COIN could help improve on DLT realizations; all activities resulting from
research on topics I see as relevant to and within COIN.
> >
> > So this gives already three agenda items from where I’m coming from
(depending on willingness for time allocation, between about 45 to 60mins
on an agenda in my mind) but yet we are told at ‘we cannot put a good
agenda together’. Is there nothing beyond these items, really, and/or is
this a judgement of those items in quality (I would expect good discussions
on them but maybe it is just me)?
> >
> > So I’m disappointed but also shocked by this style of simply cancelling
the RG meeting with that (too) thin ‘we cannot put a good agenda together
for IETF113’ explanation. I cannot and do not see the reasoning behind it
albeit I may speculate but I am not a friend of those second guesses.
> >
> > Hence, I would ask the community here: what discussions were we looking
forward to have? Are those good enough to discuss regardless of the RG
meeting being cancelled? If there is no RG meeting for whatever reason,
maybe we can simply come together among those interested in those
discussions and have them regardless, such as in a side meeting of the
‘COIN community’ (not the RG)?
> >
> > From my side, I would be highly interested in that since I have valued
the COIN discussions over the past years and don’t want to let go of this
for reasons that are just not well enough explained below.
> >
> > Best,
> >
> > Dirk
> >
> > From: Coin [mailto:[e-mail]] On Behalf Of Marie-Jose
Montpetit
> > Sent: 11 March 2022 00:45
> > To: coin <[e-mail]>
> > Cc: coinrg-chairs <[e-mail]>
> > Subject: [Coin] Cancelling the COINRG meeting at IETF113
> >
> > Dear all:
> >
> > Because of many converging issues, delays and (non) availability of
invited researchers and papers we cannot put a good agenda together for
IETF113. Hence we are cancelling the meeting.
> >
> > We plan to re-group, consult the community and plan for 114.
> >
> > Discussions on the use cases and other important COIN topics will have
to continue or be initiated on the list for now. Of course as the co-author
of a draft that was going to be presented I am disappointed.
> >
> > The co-chairs are in full agreement that this is the right decision at
this point and the IRTF leadership has been kept in the loop.
> >
> > J/E/M
> >
> > Marie-José Montpetit, Ph.D.
> > [e-mail]
> >
> >
> > --
> > Coin mailing list
> > [e-mail]
> > https://www.irtf.org/mailman/listinfo/coin

> --
> Coin mailing list
> [e-mail]
> https://www.irtf.org/mailman/listinfo/coin


-- 
---
[e-mail]

-- 
Coin mailing list
[e-mail]
https://www.irtf.org/mailman/listinfo/coin