[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [kNT] Multicasting [was -- akamai]
Hi Scott,
You're absolutely right about the technical issue... and it is a big
problem for some network providers to roll multicast into the
backbone... But I also truly believe that no one will see the point in
paying a bunch of guys (who are going to be pretty expensive) to manage
and roll out a multicast enabled IP backbone project when, today, there
is little benefit. Don't get me wrong though, it has been done, in fact
there are over 20 networks that peer at the LINX using multicast today.
So I think if there was a real need there to do it, it would be done,
but cash is tight now and it's not really a priority.
As far as people losing money goes, again, you're totally right. I went
to an interesting seminar setup by linx which was about how ISPs will
make money from multicast... and I came out a lot less clear than when I
went in! Needless to say the model is far from complete.
On the "making content multicastable"... I'd had a few beers last night
so probably wasn't making my self clear. The content, you are right IS
irrelevant and in fact the server side of the process again is very
easy. It's as easy as ticking a box in Real server and typing in a
multicast group (well almost) and in enterprise networks they already
use multicast for streaming media etc. What I was trying to get it is
that the model used for multicasting over the internet at the moment has
some problems. One of the key issues being the multicast group
addressing scheme which is something that source specific multicast
(SSM) will help to over come.
Its funny you should ask about the multicast streams but the BBC
actually had some going at some point! I think world news 24 or whatever
it's called was getting multicasted. You should speak to Simon
Lockheart...
Multicast wouldn't just be beneficial to media based content... ICQ gets
downloaded about a million times in a bad week... last time I looked the
file is about 5mb in size... If you do the maths that uses up quite a
lot of bandwidth... I'm sure caching helps to a certain degree. If you
were to distribute this type of content using multicast with a constant
stream of about 6mbits from the server you could cater for millions and
millions of simultaneous downloads and the best bit is the speed would
be very quick as the content is actually coming from the "wire" rather
than a source server or even an edge server. And for the provider a half
decent server could deal with ALL the traffic no more load balanced
server farms.
No doubt in my mind that multicast IS the way forward, technically
anyhow.
Pro multicast rant ends here :)
Nick
-----Original Message-----
From: owner-k-networktrial@xxxxxxxxxxxxxxxxxxx
[mailto:owner-k-networktrial@xxxxxxxxxxxxxxxxxxx] On Behalf Of Scott
Pollard, BIBC
Sent: 09 November 2001 10:13
To: k-networktrial@xxxxxxxxxxxxxxxxxxx
Subject: Re: [kNT] Multicasting [was -- akamai]
Well, it *is* getting a little bit technical... Daniel, maybe there's
scope for a 'techie' list?
But... isn't the problem with multicast two-fold? First one is
technical -- all intervening routers need to be multicast enabled, and
if the operator doesn't control the end-to-end network then that might
be [is] difficult to co-ordinate. Second is financial -- if the
operator is paid by connectivity required [megabits/second per month],
or gigabytes transferred, what is their incentive to reduce the amount
of traffic. After all, we're hearing there is a glut of fibre and so
sending 1000 identical streams of 300 k each is a good way of being paid
to fill it... isn't it?
Well, no it's not.
a] because it is so manifestly inefficient and unreliable that content
owners are looking for something better to ensure reliable service they
can charge for
b] operators themselves have to pay for, or lay, their own dark fibre to
be filled with these identical streams so that reduces their margins
I believe that some of the more imaginative operators are implementing
multicast and mcast peering but I suspect it will be a long while before
it gets really close to the 'last mile' router.
But I don't understand Nick's comment about "making content
multicastable" -- surely the actual content is irrelevant as long as it
is a stream [like a TV channel] rather than VoD?
Does any content owner/broadcaster have a live multicast operation in
use 24*7? I would be very interested in making contact.
Scott
----- Original Message -----
From: Nick Bryant
To: k-networktrial@xxxxxxxxxxxxxxxxxxx
Sent: Thursday, November 08, 2001 11:57 PM
Subject: RE: [kNT] akamai
Ya see Marshall,
I got to disagree. MSDP PIM and MBGP are all working fine? but to be
honest until BGMP and IGMPv3 with PIM-SSM are widely deployed then its
not really the content your waiting for? it?s the ease of making
content multicastable ? that is holding everyone up. There is plenty of
content there? kinda like the chicken and the egg?
Just my thoughts anyhow?
Nick
PS is this a ?geek? group and should be subjecting people to this?
-----Original Message-----
From: owner-k-networktrial@xxxxxxxxxxxxxxxxxxx
[mailto:owner-k-networktrial@xxxxxxxxxxxxxxxxxxx] On Behalf Of Marshall
Eubanks
Sent: 08 November 2001 17:40
To: k-networktrial@xxxxxxxxxxxxxxxxxxx
Subject: Re: [kNT] akamai
John McHugh wrote:
Nick,
What you say here is uncannily very close to the truth! Wait and see
what
AOL/TIME/WARNER comes out with in the near future, another idea stolen
from
the pram of youth.
Multicast is something I've been around since 1987 and I think there is
something we don't know....... like big brother has always kept hold of
the
last bit of the jigsaw so at the end he can say "look I finished it" and
then declare that it was just a bespoke version of Quake XIIVI and was a
test to see how far we got?
Realistically as the consumer connectivity builds then I think so will a
multicast peering model. What's required is an early leader in the
subscription based revenue space to be "got on board" and then just
leverage
the rest with money! (by the way this is in the vernacular in case you
thought I was a bread head)
John
Or even better still a multicast peering solution. Media would then no
longer be an issue and some very clever people could write some
protocols for distributing bulk data reliably over a multicast
infrastructure. Or maybe we already did.... ;)
But I've given up the whole multicast thing now as it seems to be making
sod all progress... unless someone knows something I don't?
Nick
Dear John, Nick;
Multicasting is far from stagnant - in fact, it is more and more seen
as the way to stream
cost-effectively. Particularly for video streaming (where a 2 hour movie
might cost $5
per copy to stream on the commodity Internet at a reasonable bit rate),
multicasting is
seeming more and more compelling.
On a technical side, things have settled down a lot. PIM-SM, MBGP &
MSDP are
widely deployed and work pretty well. People are also much less afraid
of multicast peering than
they were a few years ago. The biggest thing that is missing is content.
--
Regards
Marshall Eubanks
T.M. Eubanks
Multicast Technologies, Inc
10301 Democracy Lane, Suite 410
Fairfax, Virginia 22030
Phone : 703-293-9624 Fax : 703-293-9609
e-mail : tme@xxxxxxxxxxxxxxxxx
http://www.on-the-i.com
Test your network for multicast : http://www.multicasttech.com/mt/
Check the status of multicast in real time :
http://www.multicasttech.com/status/index.html