LISTSERV mailing list manager LISTSERV 16.5

Help for AFRIK-IT Archives


AFRIK-IT Archives

AFRIK-IT Archives


AFRIK-IT@LISTSERV.HEANET.IE


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

AFRIK-IT Home

AFRIK-IT Home

AFRIK-IT  September 1997

AFRIK-IT September 1997

Subject:

Re: Access <== Cost

From:

Nemo Semret <[log in to unmask]>

Reply-To:

African Network of IT Experts and Professionals (ANITEP) List

Date:

Mon, 29 Sep 1997 16:25:55 -0400

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (123 lines)

Greetings all,

Christopher Geerdts (in the posting included below) raised the key
issue in peering between ISPs/NAPs. Traditional peering has a problem
when there are asymetric flows, i.e. a small ISP benefits greatly by
getting a "free ride" onto the backbone whereas the larger one gets
only marginal benefits from being able to reach the small ISP's user
base. In the US, this has prompted the large ISPs who are also
backbone providers like MCI, UUnet (now owned by Worldcomm) to
introduce new pricing policies.

Indeed it is not interconnection itself, but peering without
appropriate pricing that has problems. As the number of users grows,
upgrades in backbone equipment are required to keep the quality
reasonable. All the added profits from new users go to the ISP they
join, but with free peering, not enough of it goes to the network
access providers, where a large part of the added cost is (as
Christopher pointed out). Thus, the access providers have insufficient
incentive to upgrade the network, and everyone suffers from more
congestion. Or they jack up their prices, and the smaller ones have to
cough up the $ or go somewhere else. There's no free lunch (at least
not forever).

It must be noted that if ISPs in Kenya, for example, do not peer
locally, they are not avoiding this problem. They are simply pushing
it out to Europe or the US, where they are even smaller players, and
more vulnerable. Better to do it locally/regionally, where you are a
larger part of the market and can negotiate a fair deal, or as I
suggested in my previous posting, you are a partner with others in
operating the access point itself.

As Internet use grows, the "fair deal" is evolving from free peering
toward interconnection with prices reflecting the traffic flows. UUnet
recently announced that they will freely interconnect only with
providers who have a national (in the US) backbone network of at least
45 Mbps, and can interconnect with them at four geographically
widespread points. The others (smaller ISPs) will have to pay to
connect to UUnet, just as a corporate intranet has to pay to connect
to the Internet.

Eventually, this can get very sophisticated, with traffic monitoring,
policing, automated auctions (my favourite), using a variety of tools
now in the reasearch arena but soon to be part of the equipment and
protocols. In the long run, I believe we will evolve a global dynamic
market where bandwidth is traded like wheat, oil and other essential
commodities are today.

Christopher Geerdts concluded thus: "peering and bandwidth sharing is
good for the consumer, but ISPs need to perceive a benefit to outweigh
the effort and potential problems".

Indeed!  The smart ISP is the one who realizes that what's could for
the consumers is good for the provider. Remember, this should be a
growth business, especially in Africa. To reiterate my main point,
local interconnection gives a huge increase in efficiency of bandwidth
use, because of the multiplexing gain. It will be *the* key to
reducing cost and increasing end-user performance dramatically, as
long as: a) 10000km undersea optical fibers or satellite connections
remain expensive, and b) that's where the bandwidth bottlneck is most
likely.  The related pricing issue is a very real one, but it's a
problem only if you try to run away from it. I see it more as a great
opportunity.

For more info, here's some related stuff (courtesy of the hotbot
search engine):

http://www.clark.net/pub/rbenn/debate.html
http://www.techweb.com/wire/news/may/0512uunet.html

        -nemo-

---
http://www.ctr.columbia.edu/~nemo


> I'm replying to a the one-week old posting on the idea of ISPs
> sharing bandwidth in Kenya.
>
> Sharing bandwidth appears to be an obvious way forward, but has it's
> problems, which (if you think about it) is the most likely reason why
> the ISPs are not actually doing it:
>
> Let's take peering as an example - where packets from one
> ISP to the other cross the peering link within the country, rather
> than going to the US and back.
>
> This works well as a bilateral arrangment between two similar-sized
> ISPs. The benefits are obvious. Only Friday afternoon, I could not
> access my POP server between 2 cities in South Africa (even though I
> was the only one logged onto a 128K dedicated line), because the ISP
> I was accessing from did not have a peering link to the ISP with my
> mailbox server.
>
> When ISPs are unequal, the smaller one stands to gain, since its
> smaller base of users can now gain faster access to a larger base of
> users. Within the larger base, the bulk of the traffic is
> international traffic anyway (most traffic is web-browsing, and
> inward bound packets). The next largest proportion of traffic is
> internal to that ISP anyway, since it has a high share of users. Only
> a small proportion of traffic crosses the peering link.
>
> If the one ISP is much larger, it will rather not peer, as that will
> mean more effort, small gain and losing one of its competitive
> advantages.
>
> When there are more ISP's involved, new issues arise, and the whole
> thing gets even more complex. Of course, in multi-peering
> arrangements, it's easier for a small ISP to catch a free ride on the
> other ISPs' international bandwidth, thus saving on the biggest
> operational cost! It's also more difficult for the individual ISP to
> promote it's link as the fastest.
>
> In short, peering and bandwidth sharing is good for the consumer, but
> ISPs need to perceive a benefit to outweigh the effort and potential
> problems.
>
> Christopher Geerdts
> International Development Research Centre
> phone: +27-11-3391911; fax: +27-11-3395050
> mailto:[log in to unmask]
> http://www.idrc.org.za/connectsa
>

Top of Message | Previous Page | Permalink

Advanced Options


Options

Log In

Log In

Get Password

Get Password


Search Archives

Search Archives


Subscribe or Unsubscribe

Subscribe or Unsubscribe


Archives

January 2012
December 2011
November 2011
March 2011
February 2011
January 2011
September 2010
June 2010
May 2010
April 2010
March 2010
February 2010
October 2009
September 2009
July 2009
June 2009
March 2009
February 2009
January 2009
November 2008
October 2008
August 2008
July 2008
April 2008
March 2008
November 2007
August 2007
July 2007
February 2007
January 2007
December 2006
November 2006
October 2006
September 2006
August 2006
July 2006
June 2006
May 2006
April 2006
March 2006
February 2006
January 2006
November 2005
October 2005
August 2005
July 2005
June 2005
May 2005
April 2005
March 2005
February 2005
December 2004
November 2004
October 2004
September 2004
August 2004
July 2004
June 2004
May 2004
April 2004
March 2004
February 2004
January 2004
December 2003
November 2003
September 2003
August 2003
July 2003
June 2003
May 2003
April 2003
March 2003
February 2003
January 2003
December 2002
November 2002
October 2002
September 2002
August 2002
July 2002
June 2002
May 2002
April 2002
March 2002
February 2002
January 2002
December 2001
November 2001
October 2001
September 2001
August 2001
July 2001
June 2001
May 2001
April 2001
March 2001
February 2001
January 2001
December 2000
November 2000
October 2000
September 2000
August 2000
July 2000
June 2000
May 2000
April 2000
March 2000
February 2000
January 2000
December 1999
November 1999
October 1999
September 1999
August 1999
July 1999
June 1999
May 1999
April 1999
March 1999
February 1999
January 1999
December 1998
November 1998
October 1998
September 1998
August 1998
July 1998
June 1998
May 1998
April 1998
March 1998
February 1998
January 1998
December 1997
November 1997
October 1997
September 1997
August 1997
July 1997
June 1997
May 1997
April 1997
March 1997
February 1997
January 1997
December 1996
November 1996
October 1996
September 1996
August 1996
July 1996
June 1996
May 1996
April 1996
March 1996
February 1996
January 1996
December 1995
November 1995
October 1995
September 1995
August 1995
July 1995
June 1995
May 1995

ATOM RSS1 RSS2



LISTSERV.HEANET.IE

Secured by F-Secure Anti-Virus CataList Email List Search Powered by the LISTSERV Email List Manager