[Chapter-delegates] On the Bandwidth congestion and the inevitable internet crisis.

Fri Mar 28 09:05:52 PDT 2008

Dear Chris Grundmann,

I agree that my observations of the unknown environmental dimensions
distracted the main topic. So the focus here is on the bandwidth
congestion problem. I have also taken the liberty to change the
subject line to keep the core discussion in focus.

I followed the link and read Mike Ou's article as also about his visit
to Capitol Hill. This is what I feel.

The present TCP control works like a free for all and laneless bridge
for automobile traffic over a wild river. The TCP control dumps a few
automobiles onto the river when there are too many cars side by side.
Partial roadblocks on either side of the bridge go up if there are too
many cars seen thrown into the river. ( The analogy is on the
fundamentals of the problem - it is not exactly relevant in various
details. It is not to be misunderstood as advocacy for "control" of
the Internet )

1.  The traffic congestion needs to be addressed seriously.

2.  a) Creating more capacity and b) managing networks are the
solutions but both these solutions need to be carried out
simultaneously, Creating more capacity alone does not solve the
problem as it is evident from Japan's experience with 100 MBPS
dedicated fiberlink to homes. With such infrastructure, clogging or
slow-down is an issue in Japan. It is reported that "1% of the users
primarily through P2P consumed around 50% of the total capacity"
Elsewhere, even a developed nation like the US is way behind the 100
MBPS standard of Japan. In the US when one talks about increasing
capacity, relatively lower bandwidth is talked about ( for e.g DOCSIS
3.0 is about providing 120 MBPS shared bandwidth between a hundred or
more users ).  If the P2P proponents continue to argue against any
form of rationalization, such increase in capacity is not going to
solve the clogging problem, not even temporarily. So it is necessary
to simultaneously manage the networks while working on capacity
increases.

3. Most arguments on net neutrality w.r.t. to the P2P traffic
restrictions by ISPs are  misguided. P2P proponents are unfair in
claiming that the ISPs are unfair, because it is P2P that is
technically unfair. P2P came into existence with an exploit of the
loopholes. The TCP protocol with Van Jacobson's patch did not
anticipate multiple streams with multiple persistence so there were no
laws or even conventions on this aspect. P2P was invented in the
absence of such conventions on the number of streams and the intensity
of persistence. Technically speaking is founded on a trick.

4. I am not against P2P. But am simply against the idea that P2P
should oppose any attempt by anyone to even balance the P2P traffic.
It is reported that "fewer than 10% of all Internet users using P2P
hogs roughly 75% of all network traffic". If this isn't unfair, the
measures by ISPs to restrain P2P traffic isn't also unfair.

5. There are two kinds of P2P users. The first kind is that of an user
who can't wait for 40 minutes for a 4 GB movie to download. He uses
P2P because it provides a TECHNICAL solution. The other kind is that
of a user who requires tremendous bandwidth, at someone else's cost,
so he uses P2P for financial reasons. For argument's sake, if the ISPs
drop all their objections to P2P traffic in exchange for
proportionally additional subscription payments by the multi stream
usage with multiple persistence, user one should have no further
complaints. His technical need for P2P is taken care of. But user 2
will continue to scream foul.

6. Bandwidth capacity can be increased in some geographic zones and
can not be increased in a few other zones. For instance WiFi has some
limitations There is a scarcity of Wireless spectrum, and it is argued
that if P2P clogs wifi bandwidth, the wifi dependant rural and remote
users (who don't have a wired Internet option) would get dropped from
Internet and even the telephone.

7. Such considerations makes it necessary to consider ways of
separating non-intensive uses from the intensive uses. In the same
thread - with the earlier subject line Article in New York Times - I
have raised a question " What if all video and audio is moved to the
TV screen by [a separate] cable ? " . What are the technical ways of
doing this ? If it is possible to have altogether separate medium /
infrastructure for intensive uses such as a HD movie download, those
users might truly share the infrastructure dedicated for such uses.
Such a measure could ensure that bandwidth for basic communication and
for more purposeful uses of the Internet remains always up and
available, in isolation of the bandwidth for intensive uses.

8. One of the objectionable solutions proposed are "metered Internet
access". If this is not acceptable, would it be acceptable if ISPs
move away from bandwidth plans to a subscription plan based on the
number of streams ? This ought to be fair.

The Internet Governance has to move away from the mindset that the
Internet is not to be governed. It needs to be governed at least in
the sense of bringing about a balance. But it is very very difficult
to bring about a balance, as politicization of issues is inevitable in
the process. P2P proponents see a motive on the part of the ISPs, that
the ISPs are primarily interested in killing the competition for video
distribution. It is not fair to dismiss this as entirely unfounded.
So, "balance" implies that P2P should be allowed to co-exist and grow,
with the issue of clogging sufficiently addressed.

In general, in all situations, at all times, there is a status quo.
Whenever there is some discussion on altering the status quo
resistance builds up. Various stakeholders get worried about their own
interests and lobby to direct development towards an atmosphere most
congenial for their own self interests. What is often missed is that
most of what they resist are long term changes. Most of the long term
changes, however drastic they are, would never affect the resilient
enterprise.

(Imagine - for some unlikely reason - the people of the world decide
to place severe restrictions on soft drink companies on the use of all
conceivable forms of packaging materials for use by the soft drink
industry. In this wild example, if there is going to be a ban on use
of tetrapak, bottles, pouches, PET and metal by all software companies
from year 2010, do you think Coca Cola and Pepsi Cola would disappear
as enterprises ? Their products might change, their business models
might change and for all we know these corporations might emerge
stronger and richer after such an imaginary regulation. These names
are mentioned for the simple reason of making this point easlly
understandable.)

Long term developmental measures are to debated without concern for
the next quarter. The corporate world often lives quarter to quarter,
so most of the politization of long term measures origninate from
short term concerns that are bound to dissipate in the long term.

It takes a certain degree of openness on the part of a phone company,
an ISP, a P2P corporation and an user to do what is right for the long
term.

We do need an overall blueprint for the evolution and the growth of
the Internet.

Sivasubramanian Muthusamy
ISOC India Chennai.
http://www.isocindiachennai.in
chennai at isocindiachennai dot in
+91 99524 03099

On Wed, Mar 26, 2008 at 12:00 AM, Chris Grundemann
<cgrundemann at gmail.com> wrote:
> Hello everyone!
>
> This is something of a shift in the subject of this thread but my
> interpretation of the NY Times article focused more on the mechanical issues
> of Internet congestion and less on the environmental impact.  Along that
> line of logic, I read an interesting and quite informative article last week
> on Bob Briscoe's proposed solution to Internet congestion from a
> 'mechanical' / usability standpoint.  The article is on ZD Net, here:
> http://blogs.zdnet.com/Ou/?p=1078&page=1.  In summary, he is proposing
> changing the TCP stack to use stream weighting and then leveraging explicit
> congestion notification to allow more efficient use of the currently
> available Internet bandwidth.
>
> I wonder if the discussion of congestion control and "Net Neutrality" is
> appropriate here.
>
> If so, I would ask what your opinions are on the suggestion in this article
> that this is an engineering problem which has been politicized?
>
> ~Chris Grundemann
> Interim Chair, CO ISOC
>

-- 
Sivasubramanian Muthusamy
CEO
Isolated Networks
Whitefield, 389/1 Perundurai Road
Erode 638 011
Tamilnadu India
http://www.isolatednetworks.com
email: isolatedn at gmail.com
++91 424 4030334
Mobile Phone number +91 99524 03099
++91 424 4030334

DISCLAIMER:

This message (including attachment if any) from Isolated Networks is
confidential and may be privileged. If you have received this message
by mistake please notify the sender by return e-mail and delete this
message from your system. Any unauthorized use, disclosure or
dissemination of this message in whole or in part is strictly
prohibited.