site Search:


 
   
story category
AT&T Backbone Sees 20% P2P Drop
Hulu, YouTube and other alternative video 'taking over'
by Karl Bode Thursday 31-Jul-2008 tags: prices · business · bandwidth · telco · networking · AT&T U-Verse
Industry analyst Dave Burstein has an interesting (but margin blown) post over at the interesting people listserv discussing the reality of congestion (or lack thereof) on AT&T's network. While industry lobbyists use P2P congestion as a bogeyman to justify all manner of policy, AT&T data suggests P2P is actually declining on AT&T's network. Upstream P2P on cable networks remains a capacity problem, but it's one that may be resolved by a migration to DOCSIS 3.0. Burstein suggests the debate over throttling is all but dead:

Easily a third of AT&T's downstream traffic is now "web audio-video," far more than p2p and the gap is widening rapidly. Hulu and YouTube are taking over, while p2p is fading away on DSL networks. One likely result is that managing traffic by shaping p2p is of limited and declining use, perhaps buying a network 6 months or a year before needing an upgrade. The p2p traffic shaping debate should be almost over, because it simply won't work very much longer.

AT&T writes off that decline in P2P use as a statistical anomaly created by a heavy mix of new customers who don't use P2P. Still, it suggests that P2P isn't quite the network demon it's often painted as. AT&T says that as of June, AT&T traffic was about 1/3 Web (non video/audio streams), 1/3 Web video/audio streams, and 1/5 P2P. Most interestingly, Burstein suggests that capacity upgrades should more than handle growth, without throttling or raising capex, while actually lowering AT&T's per bit cost per user. On upgrades:

AT&T has sensible plans to handle the load without disruption. They are already moving from 10 gig to 40 gig in the core, and planning a transition to 100 gig in a few years. The current projections are they can do these upgrades without raising capex, bringing per bit costs down along a Moore's Law curve and keeping bandwidth costs per user essentially unchanged.

So if the capacity costs of keeping pace with demand are nominal, does that still make AT&T's push into metered billing "inevitable?" One gets the feeling that there's no greater chasm than the one between a lobbyist and network engineer describing the same network.

view: topics flat text 
Post a:
moonpuppy

join:2000-08-21
Glen Burnie, MD

Most new people do not know about P2P

Had a friend of mine that was the same way.

He heard of something called Kazaa but had no idea what it was. I tried to explain to him what P2P was and he was still in the dark. All he knew was he could get music and porn.

He knew nothing about the newsgroups until another friend of ours told him about it. He still did not understand what was behind it.

Most people are going to go to these other sites because it is easier. That's the only reason.

jester121
Premium
join:2003-08-09
Lake Zurich, IL

Re: Most new people do not know about P2P

Plus it's been a slow month for Linux kernel updates.
cybercrimes

join:2003-12-24
Honey Brook, PA

Just Upgrade

alot of isp's just dont want to spend the time and money to
upgrade there pipes and server to handle the load of users

espaeth
Digital Plumber
Premium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2
Reviews:
·Vitelity VOIP

Re: Just Upgrade

said by cybercrimes:

alot of isp's just dont want to spend the time and money to
upgrade there pipes and server to handle the load of users
That's not entirely accurate. Many ISPs are waiting for improvements in their delivery technology of choice (*DSL, DOCSIS) to turn into product so they can upgrade their network. You cannot add capacity using product that doesn't exist.

Zeke

@prioritynetworks.net

hmmmm

Do they still have the same amount of overall backbone traffic or is that declining? Maybe they are just dropping in bandwidth or customers.
axus

join:2001-06-18
Washington, DC

Re: hmmmm

Yeah, I wondered that. The data is shown as percentages (well, fractions). Has the total amount of traffic increased while the P2P traffic remained the same? Or has P2P usage actually gone down? Maybe the pirates have downloaded everything they wanted already ;p
deadzoned
Premium
join:2005-04-13
Baton Rouge, LA

1 edit

Excuses, Excuses

That's all this stuff is because the real issue with the broadband providers is NOT about bandwidth capacity as they are saying it's really about CONTROL and nothing else. They seek more control over their network and the way to assume that control and assert it is through stuff like Bandwidth Caps, Metered Billing, Throttling, etc... It's really all part of their plans to have more control.

There is no "Bandwidth Apocalypse" due to P2P, Piracy, or any other bogeyman they may trot out. It's simply mis-direction plain and simple.

insomniac84

join:2002-01-03
Schererville, IN

Re: Excuses, Excuses

It's all about needing to keep profits always increasing to keep investors happy. When you can't rely on new customers to bring in extra profits, your only option is to milk existing customers to make up the gap.

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

This is alarming

20% of all traffic consist of illegal activity.

All those users lost to Comcast must be the heavy users. Good news for the rest of the legit users.

aciddrink

join:2000-08-26
Lexington, KY
Reviews:
·Insight Communic..
·Windstream

Re: This is alarming

said by ninjatutle:

20% of all traffic consist of illegal activity.
90% of all statistics are made up on the spot.

hopeflicker
Capitalism breeds greed
Premium
join:2003-04-03
Long Beach, CA
kudos:1

Re: This is alarming

said by aciddrink:

said by ninjatutle:

20% of all traffic consist of illegal activity.
90% of all statistics are made up on the spot.
and 50% of users pull numbers out of their ass without posting links to their claims.
--
Religion does three things quite effectively: Divides people, Controls people, Deludes people.
cornelius785

join:2006-10-26
Worcester, MA

Re: This is alarming

with 98% confidence that 90% of the people reading will believe it

AVD
Respice, Adspice, Prospice
Premium
join:2003-02-06
Onion, NJ
kudos:1

There are

There are lies, dammed lies and statistics...

backfeed
is giving feedback

join:2002-12-16
Peru, IN

Re: There are

Mark Twain??

NetAdmin1
CCNA

join:2008-05-22

Re: There are

»en.wikipedia.org/wiki/Lies,_damn···atistics

Benjamin Disraeli... Someone who obviously failed stats.
--
---
Eleven years of carrying The Clue Bat...

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

1 edit

Re: This is alarming

said by hopeflicker:

said by aciddrink:

said by ninjatutle:

20% of all traffic consist of illegal activity.
90% of all statistics are made up on the spot.
and 50% of users pull numbers out of their ass without posting links to their claims.
RTFA

"about 1/3 Web (non video/audio streams), 1/3 Web video/audio streams, and 1/5 P2P."

hopeflicker
Capitalism breeds greed
Premium
join:2003-04-03
Long Beach, CA
kudos:1

Re: This is alarming

said by ninjatutle:

RTFA

"about 1/3 Web (non video/audio streams), 1/3 Web video/audio streams, and 1/5 P2P."
And you believe what they say?
--
Religion does three things quite effectively: Divides people, Controls people, Deludes people.

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

1 edit

Re: This is alarming

IDK, ask Karl Bode See Profile, he was the person who posted it. Why don't you question his character?

hopeflicker
Capitalism breeds greed
Premium
join:2003-04-03
Long Beach, CA
kudos:1

Re: This is alarming

Ummm, ATT claims this:

AT&T says that as of June, AT&T traffic was about 1/3 Web (non video/audio streams), 1/3 Web video/audio streams, and 1/5 P2P.

so i ask again, do you believe this?
--
Religion does three things quite effectively: Divides people, Controls people, Deludes people.

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

Re: This is alarming

Again, I will refer you to Mr. Karl Bode See Profile if you feel his reporting is inadequate or the integrity of his linked reports are inaccurate.

This is going in circles.....

FLengineer
CCNA, CEH, MCSA
Premium
join:2007-06-26
Leesburg, FL

1 edit
That sounds about right to me. But, I'm lost in the convo here, are you saying that P2P = illegal?

EDIT: Question is to ninjatutle

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

1 edit

Re: This is alarming

I'm not buying the endless Linux disto downloads, World Of War patches and home movies to grandmother so yes, ILLEGAL material is being passed.

You know very well as with everyone here, what P2P is mainly used for.

EDIT: Reply to Flenginear

pspcrazy
Anime Freak

join:2008-02-06
San Diego, CA

Re: This is alarming

Well what about Anime which is around 95 percent of my bittorrent downloads. Do you consider those illegal? Anime companies in the U.S can't touch anything other then the ones they licensed. They are not illegal unless licensed in the u.s thus not making them illegal. Then again each anime fan has his own idea on it.
axus

join:2001-06-18
Washington, DC

Re: This is alarming

Actually I believe we have some treaties with other countries protecting each others' copyrights:

»en.wikipedia.org/wiki/List_of_pa···treaties

The anime companies just don't have an RIAA running around suing people and spreading the word that there's free stuff on the 'net. Probably to their benefit!
TheWizardhfl

join:2004-08-05
See, you are a little misguided there. Anime when produced in japan is internationally copyrighted. And yes, they have pursued methods to put an end to some of the distribution. Media Factory Inc. is well known for their issuing of C&D orders to anime "fansubbers" and Bandai has issued warnings publicly in the past regarding specific title releases.. For the most part, anime fansubs tend to rest in a gray area. The majority of production companies simply choose to overlook the issue for various reasons ranging from lack of resources to fear of alienating their consumer base.

When an anime is licensed in the US, the most notable change in regards to the topic is that the US company who licensed the title locally has the right to file C&D orders and lawsuits against anyone distributing unauthorized releases of their titles online.

It's not about who has their own idea on the matter or not. It's a matter of it's still illegal either way. The only difference is who has the right to pursue the matter legally.

FLengineer
CCNA, CEH, MCSA
Premium
join:2007-06-26
Leesburg, FL

1 edit
P2P = Illegal ??

Yes some of it is, some of it isn't. I bet you also think all Muslims are terrorist, too.

»www.bittorrent.com/
»revision3.com/
»beta.legaltorrents.com/
May Fav »torrentfreak.com/sundance-winner···torrent/
That's a good movie and was released on P2P by the producers.

You say all P2P is Illegal, I just provided proof that P2P is NOT Illegal. Revise your statement to Sharing Copyrighted material VIA P2P is Illegal.

EDIT: You mentioned World of Warcraft being insignificant.
2.5M players in North America.
2.4.0 patch distributed via P2P a week in advance was 256.11MB
That's 91.5TB A DAY

Dogfather
Premium
join:2007-12-26
Laguna Hills, CA

Re: This is alarming

Take one look at what the top search results are on BT search engines and you'll see that the mass of traffic is illegal. And so what if it is. Before BT it was Kazaa, before that Napster, before that Usenet...and after BT it will be something else.

The ISPs and content creators will NEVER defeat piracy and in their efforts all they do is piss off legit users of their services or products.

Matt3
All noise, no signal.
Premium
join:2003-07-20
Jamestown, NC
kudos:12
said by ninjatutle:

You know very well as with everyone here, what P2P is mainly used for.
That is an unreasonable, tiresome, non-factual based argument. We all ASSUME we know what it is being used for. Just like up until this article was posted, you probably would have ASSUMED that P2P was the majority of internet traffic.

Until a study is done that proves that the majority of P2P traffic is illegal, I simply can't believe it and spewing the corporate line based on an assumption with no factual basis is asinine.

Too many other legit services use P2P. Hell, I just downloaded about 7GB of WoW downloads this past weekend for my neighbor via P2P.

The FACT that P2P use is declining while YouTube and Hulu (non-P2P video based distribution methods, just like Netflix online) are rising would lead one to believe that P2P video distribution services like Joost are losing viewers. Hulu is a much better product than Joost and the other P2P TV distributors.
nasadude

join:2001-10-05
Rockville, MD
Reviews:
·Verizon FiOS
said by ninjatutle:

...
You know very well as with everyone here, what P2P is mainly used for.
Well, that's the "conventional wisdom" that the media companies and ILECs have been able to push into place. I have never seen any data to back this up, but if you scream something loud enough over and over and over, it can't help but sink into the consciousness of those not paying much attention (like the press, congress and 98% of the public).

and now there is evidence (although not trusted because it goes against the "conventional wisdom") that P2P use is declining - woe unto the ISPs! what will they use to scare us about bandwidth scarcity now?

ssj4android
Redefining Reality

join:2002-04-14
Wyoming, MI
What is counted as "peer to peer" traffic though? I'd think a majority of online gaming traffic is peer to peer, and that certainly isn't illegal.

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

Re: This is alarming

»en.wikipedia.org/wiki/Peer-to-peer
»en.wikipedia.org/wiki/Game_server

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:6
said by ssj4android:

What is counted as "peer to peer" traffic though? I'd think a majority of online gaming traffic is peer to peer, and that certainly isn't illegal.
I'm not sure what ninja was intending to communicate. Peer-to-peer is an architecture, another common architecture is client-server. Many IM and chat clients, online games, and VOIP applications make peer-to-peer connections.

Peer-to-peer is also often a verbal abbreviation of "Peer-to-peer file sharing." I think the statistics of 1/5th being P2P probably refers only to file sharing. The devices that measure such things don't really check to see if an end point is a server, instead they actually sniff the packets as they go by and figure out what known application protocols are being used within them.

So your gaming traffic is probably not included in the statistic, regardless if the architecture is peer-to-peer or client-server. If your game uses BitTorrent to do its updates, then only that part is counted in the "P2P" column and the rest of the data (move, aim, fire, character) would not be.
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More fun, more features, Join BroadbandReports.com, it's free...

Reck Havoc
Premium
join:2002-07-31
Grand Rapids, MI
said by ssj4android:

What is counted as "peer to peer" traffic though? I'd think a majority of online gaming traffic is peer to peer, and that certainly isn't illegal.
Right, little ninja buddy thought it'd be clever to link wiki articles.. But he forgets fact, as seen through all his posts.

The entire Xbox 360 and most of the PS3 online games use a P2P network, where either one person is actually hosting the server, or where it's split between all the users upload equally.
--
AMD Athlon 64 FX-53, 1GB Corsair XMS Dual Channel PC-4400 DDR,NVIDIA GeForce 6800 Ultra 256MB, 160GB Seagate SATA - RAID 0, Windows XP Pro

Nightfall
My Goal Is To Deny Yours
Premium,MVM
join:2001-08-03
Grand Rapids, MI
Reviews:
·Comcast
·Callcentric
·Site5.com

Re: This is alarming

said by Reck Havoc:

said by ssj4android:

What is counted as "peer to peer" traffic though? I'd think a majority of online gaming traffic is peer to peer, and that certainly isn't illegal.
Right, little ninja buddy thought it'd be clever to link wiki articles.. But he forgets fact, as seen through all his posts.

The entire Xbox 360 and most of the PS3 online games use a P2P network, where either one person is actually hosting the server, or where it's split between all the users upload equally.
Link? Source?

ssj4android
Redefining Reality

join:2002-04-14
Wyoming, MI

1 edit
Yeah, that's what I was referring to. Wii as well, which is the only system I've captured traffic for. Brawl seems to have one main game server, but in a three player game each player sends UDP packets to the other two.
Indeed, most P2P file transfer applications use TCP, which is connection based and requires one node to be a server while the other is a client. Gaming usually uses UDP, which seems more like a "peer to peer" transport protocol to me, but does often use at least temporary host servers. In games like Halo 3, all peers maintain enough state information to become the host if the current one drops out. AFAIK, voice communication on Xbox LIVE is direct from one peer to all others.

There's also programs such as Joost. What is that counted as?

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:6

1 edit

Re: This is alarming

said by ssj4android:

most P2P file transfer applications use TCP, which [...] requires one node to be a server while the other is a client.
I'm not sure what you mean here, but either way I work it out its inaccurate. TCP doesn't require one side to be server and the other to be client, and the very definition of P2P means that there is no server and no client. It may be a distinction without a difference, but it hit me funny.
Gaming usually uses UDP, which seems more like a "peer to peer" transport protocol to me,
Also, a case where one UDP has nothing to do with the P2P nature.

Think of TCP and UDP as languages.
TCP has already worked out rules of etiquette about who says what under what conditions, when to ask for information to repeated, and how to say hello and goodbye. TCP is fine for most uses, but sometimes it lack features you want or in the case of streaming video, it has features that you don't need.

UDP is a much looser language, where the UDP "syllables" can be used to create exactly the languages that you want. You can have the connectedness stateful nature of TCP without the retransmission request for a 10% bad packet, for example.
Think of P2P or client-server as military ranks.
At some level, all hosts on the internet are peers but some peers have a central function -- e.g. they serve web pages, or answer file-transfer requests, or respond to other queries. These hosts are servers (masters) and the hosts that connect to them are clients (supplicants). That architecture is client-server. In a client-server arrangement, clients only connect through servers who arrange all of the processing and other communication. You have to follow the chain of command.

In a peer-to-peer architecture, there is no hierarchy. End-hosts communicate with other end-hosts directly and no central device is required. If there are communal processing tasks to be performed, this work is generally distributed among the peers somehow. There is no hierarchy.
I hope that helps.
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More fun, more features, Join BroadbandReports.com, it's free...

ssj4android
Redefining Reality

join:2002-04-14
Wyoming, MI

Re: This is alarming

I know how TCP and UDP work, I just don't know what to consider a "server" (I like precise definitions).
My logic is TCP requires one node (the server) to listen for a connection and another node (the client) to initiate the connection. But perhaps that's a faulty definition, as which node was the listening one is irrelevant in, say, bittorrent.
The purest example of a p2p paradigm I can think of is a multicast chat program, where one node doesn't know and doesn't care how many other nodes are on the network.

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:6

Re: This is alarming

said by ssj4android:

I know how TCP and UDP work, I just don't know what to consider a "server" (I like precise definitions).
My logic is TCP requires one node (the server) to listen for a connection and another node (the client) to initiate the connection. But perhaps that's a faulty definition, as which node was the listening one is irrelevant in, say, bittorrent.
The purest example of a p2p paradigm I can think of is a multicast chat program, where one node doesn't know and doesn't care how many other nodes are on the network.
You are observing that most servers listen for incoming connections -- but that's common, not definitive. Who makes the connection and who listens is more of a function of how the end-points set up the communication channel, but client-server has more to do with centralization of actual work or resources.

To illustrate that it's the processing that is the chief consideration, consider the X-Server which connects outbound to X-client programs but since the terminal does the graphics processing, it is the server to the client applicaitons.

Here's one definition: »www.sei.cmu.edu/str/descriptions···ver.html
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More fun, more features, Join BroadbandReports.com, it's free...
nocannothave

join:2006-10-14
Kennewick, WA
Of course illegal material is being passed.

So are concerts of taping/trading friendly bands, tv shows, etc.

So, 20% is not illegal. 20% is p2p. That's all you can say.

Modus
I hate smartassery on forums
Premium
join:2005-05-02
us
I do. I think all the ppl that used p2p on dsl networks changed over to other isp's like comcast, verizon and etc because the speeds. ATT dsl doesn't have speed
--
Think Ahead. Learn More. Solve Now!
dentman42
Premium
join:2001-10-02
Columbus, OH

Re: This is alarming

said by Modus:

I do. I think all the ppl that used p2p on dsl networks changed over to other isp's like comcast, verizon and etc because the speeds. ATT dsl doesn't have speed
I have better bandwidth on my AT&T DSL than on my Road Runner. Even the top RR tier here only matches AT&T's Elite tier's upstream of 768k. I'm not going to pay $10 more per month for only 256k more upstream on RR.

FLengineer
CCNA, CEH, MCSA
Premium
join:2007-06-26
Leesburg, FL

Re: This is alarming

DSL vs. Cable is completely subjective to not only your market but your particular street.

ptrowski
Got Helix?
Premium
join:2005-03-14
Putnam, CT
kudos:4
Reviews:
·VOIPo
said by ninjatutle:

RTFA

"about 1/3 Web (non video/audio streams), 1/3 Web video/audio streams, and 1/5 P2P."
You RTFA, is says P2P, not illegal activity. But your trolling once again is easily noticed.

See 21 replies to this post
bjbrock9

join:2002-10-28
Mcalester, OK
Not all p2p traffic is illegal. A lot of it is legitimate file sharing. So your assumption is wrong and your numbers are wrong.

pspcrazy
Anime Freak

join:2008-02-06
San Diego, CA
With 100 percent confidence the post below me will be asking for another percentage.

Nightfall
My Goal Is To Deny Yours
Premium,MVM
join:2001-08-03
Grand Rapids, MI
Reviews:
·Comcast
·Callcentric
·Site5.com
said by hopeflicker:

said by aciddrink:

said by ninjatutle:

20% of all traffic consist of illegal activity.
90% of all statistics are made up on the spot.
and 50% of users pull numbers out of their ass without posting links to their claims.
This is 100% true. I have not seen any actual data from an unbiased source that points out the actual data usage and breaks it down factually. Instead, we have a bunch of ISPs claiming that P2P is killing them, but we have others not on the inside saying they are full of crap. So who do we believe? The majority of users here on BBR believe that the ISP is full of crap. I would like to see some real data.

The only actual data I have seen was on Comcast's network in my area. I was fortunate enough to have a friend who worked at Comcast to show me the usage data for all of the high speed internet lines in the city as broken down by usage. It was an eye opening experience to be able to see that and to see that the top 5% of the lines in the city ate up more than the last 95%.

It would be useful to see that kind of data from every ISP.

See 8 replies to this post

Doctor Four
My other vehicle is a TARDIS
Premium
join:2000-09-05
Dallas, TX
said by hopeflicker:

said by aciddrink:

said by ninjatutle:

20% of all traffic consist of illegal activity.
90% of all statistics are made up on the spot.
and 50% of users pull numbers out of their ass without posting links to their claims.
That definitely applies to the poster in question (who said
20% of all traffic consist of illegal activity).

WOOP! WOOP! WOOP!

Anti-Piracy troll detected off the port bow, Captain! Your
Orders, sir?"

"Fire main cannons, matey!"
--
"The trouble with computers, of course, is that they are very sophisticated idiots." - Doctor Who (from Robot)

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

Re: This is alarming

As with commies, the only good pirate is a dead pirate.

You missed my battleship

mathwizkid

@bellsouth.net
5/4ths of the people don't understand statistics

one_bored_si

join:2003-03-10
Montebello, CA

Re: This is alarming

lawl

ninjatutle
Premium

join:2006-01-02
San Ramon, CA

2 edits
edit:

Wrong reply.
deadzoned
Premium
join:2005-04-13
Baton Rouge, LA
What kind of Kool-Aid did you get to drink? Something good like Cherry or Grape? Hope it's good and that it is quenching that thirst of yours for corporate propaganda.

hopeflicker
Capitalism breeds greed
Premium
join:2003-04-03
Long Beach, CA
kudos:1

Re: This is alarming

said by deadzoned:

What kind of Kool-Aid did you get to drink? Something good like Cherry or Grape? Hope it's good and that it is quenching that thirst of yours for corporate propaganda.
Probably thinks that borrowing book from the library , recording an episode of the Simpsons, or loaning the new Britney Spears CD to a friend is piracy too.
--
Religion does three things quite effectively: Divides people, Controls people, Deludes people.

Bellundo

@teksavvy.com
Who cares? People who make crank phone calls from pay phones pay their 25 cents the same as someone who makes a normal phone call pays.

ninjatutle
Premium

join:2006-01-02
San Ramon, CA
A simple search on google yielded those results

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:6

Re: This is alarming

Yeah, and I'm sure you picked it because isoHunt is so heavily trafficked by legal P2P users. C'mon. Don't be lazy. Use a real source.

Actually, what if I conceded the point and said that most P2P is probably still some kind of copyright infringement, despite recent trends? So what?

beerbum
Premium
join:2000-05-06
Reading, PA

The Article - in readable format

From: Dave Burstein [daveb@dslprime.com]
Sent: Tuesday, July 29, 2008 6:10 AM
To: David Farber
Subject: Practical Congestion Solutions (Real Soon Now)

Folks

David Reed asked about how is Comcast managing traffic today, and what is likely later in the year. I have some of the data. I would enormously appreciate better information on what we can expect from DOCSIS 3.0 upstream when it's widely deployed (?late 2009.) I'm guessing it will be very good, but traffic models or other info that translate from 160/120 shared to realistic expectations per user would really help. My guess is 50/25 will be achieved 95+% of the time and that most of the cable upstream problems will be solved, but I just don't have enough data.

Takeaways:
1) The traffic at AT&T is rapidly shifting from p2p to streaming video such as YouTube and Hulu, per data below on the record and factchecked at a senior technical level. This matters, because any p2p solution will not do very much if p2p is much less of the traffic. It might postpone the need for an upgrade for 6 or 12 months, but p2p throttling can't do much more for most networks.

2) The updated model from Sandvine, Comcast's main supplier, allows much less obtrusive shaping. It can be tuned to neighborhood CMTS units rather than a wide territory, can automatically shut off when congestion clears, and enables more careful policies. Things like "if the user has uploaded more than 3 gigabytes in the last five days, reduce that connection to 256K until the congestion clears." Whether Comcast will choose a system like that, and whether the software will deliver everything promised, is not yet known. It certainly could come closer to "reasonable" than what they are doing now.

3) There really is a problem on (at least some) cable upstreams today, based on what I hear from people I respect who have the data. My hope - which won't be tested until 2009 - is that the DOCSIS 3.0 upstream will resolve most or all of the problems for the next few years. Full DOCSIS 3.0 has a minimum of 120 megabits upstream (shared) among typically 300 homes, something like 400K per subscriber. Current cable modems typically have 8 to 30K per subscriber. This is a huge difference.

Verizon, AT&T, and Free.fr are strongly on the record they do not have significant congestion problems. They do not have a shared local loop, and have allocated much more bandwidth per customer. I've sat at the network console of a large ISP and seen they had essentially no congestion amongst the millions of subscribers they served. They allocate 250K per subscriber, much more than current cable. So my best guess is that when cable gets to levels like that, the problem will drastically diminish. DOCSIS 3.0 120 meg shared downstream is now available to nearly 20 million homes (8M at J:COM Japan, millions more at Numericable France, Virgin UK, Videotron Quebec, Korea, and some others.) Very few systems can send at 50 and 100 megabits, leaving enough capacity for customers to actually receive at 50-100 megabits despite the sharing. Unfortunately, the upstream is not ready. Vendors say 2008, many think late 2009. 50/25 to 20M Comcast customers (promised for 2010) and many others is totally destabilizing if the telco doesn't have fiber and the cablecos price like they already are in Japan (100 meg is only about $5 more) and France (100 meg, shared, is effectively $25-30 as part of a bundle.) The price will be the key.

Separately, a note came from Singapore to this list as I was writing, sensibly discussing whether problems are due to the ISP or the carrier. I wish we could think like that over here, but in the U.S. and Canada something over 95% of consumers get everything from the telco or cableco. We just don't have many surviving ISPs.

---------------
I unfortunately don't have the time to integrate these longer comments with the above, so forgive the length. If I had more time, I'd write shorter. The AT&T traffic data at the end is particularly interesting; the rest is not thoroughly factchecked yet but I think mostly on target.

1) The latest Sandvine quarterly call described what they say is now available for testing. Unlike the current models, which need to do something broad like slowing all the p2p across large systems, the new software allows much finer tuning, to the level of the usage of the individual customer. It essentially is a policy server tuned to let policies be set for each customer and enforced in real time. Choice of policy would be by the carrier.

The first improvement is that it can read the network and the individual CMTS, and only invoke policy if that customer is on a congested CMTS. This dramatically cuts how much is shaped.

The second improvement is that it can choose who to shape in a much more sophisticated way. Currently, they get total usage for each customer, so implementing a 250 gig cap for an individual or grossly shaping a whole region is what they are mostly limited to (I believe). Now, they can set a policy that chooses to shape only those who use more than 4 gig upstream between 6 p.m. and 10 p.m. , and only partially shape them (say to 256K up) if that's enough. It can then turn off the shaping automatically when the circuit drops to a more normal pattern (say, less than 60% utilization for 15 minutes.)

Contrast that to today, where apparently Comcast is leaving the stuff on for 18 hours a day, and filtering closer to the core, affecting many locations without problems. (The primary problem is at the neighborhood CMTS, and the current system I believe requires shaping dozens or hundreds at a time because the boxes are deeper in the network.)

How sensible will Comcast be? I can't answer that, but since Sandvine is designing the software working closely with their large customers like Comcast, I suspect the capabilites they are discussing are what the customer intends to use.

None of which means we can let up, but suggests much of the problem can be technically fixed.

2) Neither Comcast nor any of the other cablecos has shared enough data publicly, but privately some have provided information I believe I can trust about one crucial question. There is a real problem with (at least some) cable upstream, which typically designs for 6 to 30 kbps per subscriber. p2p frequently outruns that, even if they do ordinary upgrades doubling and tripling the total shared upstream but staying within that range.

The very interesting question is whether the DOCSIS 3.0 upstream will have similar problems. Verizon and AT&T, with far more effective bandwidth per user, aren't seeing a similar problem. So my first guess is that the the p2p congestion issue is important up to a certain level of bandwidth, but not above that. This is totally unproven, and the best cable engineers I've asked are unsure until they get thousands of live customers. DOCSIS 3.0 is at least a 12x (vs 10 meg upstream cable), for many a 30x (versus 2-3 meg upstream systems.) There's a lot of folks with this "voracious p2p grabs any upstream you can give it" idea, which seems to correspond to some of the experience on cablecos today,

However, Verizon, AT&T, and Free.fr tell me they simply do not have this problem. Bell Canada provided the CRTC with data that shows similar - the only problems of significance were at the back of the DSLAM, which I've determined was they haven't upgraded the link out of the back for many years, and have many OC-3's,etc. I've sat at the network console watching the realtime network map for millions of subscribers. The only congestion at that moment was at an interconnection with Telecom Italia, where Italia refuses to upgrade their side of the link. I believe this is typical of their experience; the network engineer told me he has authority and budget to simply upgrade any link that is a problem. They have fiber throughout, all IP, and are running it as the prototype "stupid network." If they see a problem, they add another Gig-E to the fiber link.

Comcast is going all 3.0 by 2010 - 20 million homes. So one of the most important questions is whether that will be enough to avoid upstream problems. All insight welcome. I have some interesting new data points. AT&T is seeing an overall slight decline in bandwidth demand growth per user, with p2p becoming less and less of a factor. YouTube, Hulu, and other commercial streamers are growing much more rapidly.

Here's a story I have, factchecked by a VP at AT&T Labs.

20% Drop in p2p on AT&T Backbone
Other video, like YouTube and Hulu, twice as high

Easily a third of AT&T's downstream traffic is now “web audio-video,” far more than p2p and the gap is widening rapidly. Hulu and YouTube are taking over, while p2p is fading away on DSL networks. One likely result is that managing traffic by shaping p2p is of limited and declining use, perhaps buying a network 6 months or a year before needing an upgrade. The p2p traffic shaping debate should be almost over, because it simply won't work very much longer.

Jason Hillery provided some current AT&T information. p2p is currently still growing but “at a slower pace than other traffic.” On the Tier 1 AT&T backbone p2p actually dropped 20% during a period last year. AT&T Labs VP Charles Kalmanek points out that a shift in customer mix, rather than an absolute drop in overall p2p, may explain that surprising statistic. Around the world, the trend is clear: web traffic continues to grow at something like 25-40% per user each year, right in line with the trend since 2001. Video is growing rapidly, but not enough to change the trend so far.

Direct from AT&T

“Overall traffic on the AT&T IP backbone network is growing at a pace of more than 50 percent per year. This growth is a combination of customer growth and growth in traffic per customer. Average growth in traffic per customer is about 25-30 percent per year.

To gauge the application breakdown of broadband traffic, we measure downstream traffic during the weekly busy hour. With this measure, as of June 2008, traffic was about 1/3 Web (non video/audio streams), 1/3 Web video/audio streams, and 1/5 P2P (with other applications making up the remainder). For the first time in June 2008, Web video/audio was the highest traffic-generating application over our IP backbone network.

As for the trends we've seen over the past few months:
-- Web video and audio is growing at a much higher pace than overall traffic (more than 70 percent/year);
-- P2P traffic continues to grow, though at a slower pace than other traffic;
-- and Web traffic is growing at a pace consistent with overall growth.”

Many of the policy people believe that p2p is a ravenous monster that is devouring the Internet. The data show that simply isn't true, with the possible but important exception of some cable upstream. Traffic management has become so political few are willing to talk on the record, so reconciling conflicting data is very hard. It appears that large carriers are not having congestion problems on DSL or fiber backhaul, but cable upstream is. With friends, I'm trying to figure out what's going on and in particular whether the problem disappears with DOCSIS 3.0 which will have 12-30+ times more upstream bandwidth. Advice and data very welcome.

Video streamed/downloaded from sites like Daily Motion, Hulu, YouTube and others are coming to dramatically dominate web traffic. YouTube's market share is tending down, Hulu, BBC, Facebook presumably climbing. Heise has a German report that general interest in music over the web is dramatically down, consistent with my belief that most people who want music collections already have more than they can listen to. Hulu has now added movies, free with commercials. Most are old blockbusters The Fifth Element, Ghostbusters, Men in Black, Jerry Maguire and other old favorites. They have also Kurosawa's great Kagemusha, Quills, Enter the Ninja, Some Like It Hot,

AT&T has sensible plans to handle the load without disruption. They are already moving from 10 gig to 40 gig in the core, and planning a transition to 100 gig in a few years. The current projections are they can do these upgrades without raising capex, bringing per bit costs down along a Moore's Law curve and keeping bandwidth costs per user essentially unchanged. Most of the optical vendors believe they can meet those goals, although some worry that the pace of innovation may slow down as the optical components industry is struggling.
Sorry I didn't have the time to write something shorter.

db
Editor, DSL Prime

karlmarx

join:2006-09-18
iraq

Re: The Article - in readable format

So, even with Docsis 3, why are they promising 5mb/subscriber per 300 people, when they only have 120MB total? That does not make sense. Provide 1mb/subscriber, and with 300 people, that should be good enough. I mean, if you only have 1mb/sec, then you sure as hell can't use 4GB/uploads in 4 hours. Why don't they just sell what they can provide.
Do the math.
Assume 20% of their subscribers use P2P. Out of 300 houses, that would be 60 users. If every user does 5mb/sec, it would take a total of 300mb/sec to just service those users. And they already admitted they only have 120mb/sec. Solution: SELL 1mb/sec upstream! Then, those 60 users can only use 60mb/sec, and that still leaves 60mb/sec for the other 80%. Gee, very simple solution.

Oh, wait.. hmm.. Verizon provides me with 15mb/sec upstream. And they have 200 people per node, but an upstream capacticy of 1.2GB/sec. That means that 40 people at 15mb/sec use a total of 600mb/sec, and that leaves 600mb/sec for the remaining 160 people. Hmm.. which one is a better system?

Which would you choose, if the price was the same. 1mb/sec upload or 15mb/sec upload. Not a very hard decision there. 30mb/sec down or 6mb/sec down?
--
The happiest countries are the most secular. The struggle AGAINST corporations is the struggle FOR humanity!

espaeth
Digital Plumber
Premium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2
Reviews:
·Vitelity VOIP

Re: The Article - in readable format

said by karlmarx:

So, even with Docsis 3, why are they promising 5mb/subscriber per 300 people, when they only have 120MB total?
Outside of testing with Spirent, Smartbits, Ixia, or similar there is no real-world IP network scenario where all ports of attachment are maxed out.

The only time we've ever seen anything close to that on the Internet is when SQL Slammer first hit.

Your density estimate is also off a bit. For example, for Comcast they have published infrastructure stats here, they have an average of 468 homes passed per node. That's houses capable of getting a signal, but doesn't reflect actual subscriber density. They also say they have 102k HFC nodes. Last year Comcast had about 13.5 million HSI subscribers, so 13.5 mil / 102k nodes = ~132 HSI subs per node.

JasonOD

@comcast.net
So it's more of a shift from P2P to legit web video as content providers have made more available (hulu, for example), not less total traffic, just a shift. I bit of irony there I guess.

And if byte-billing doesn't work (which seems to be the thrust of the article, since it downplays capacity conerns), I smell new opportunities for ISP's. Unlike p2p, it's easy to find the bigger content houses like Hulu (again, for example). With a massive presence on ISP's networks, I'd think the ISP's would have room to negotiate revenue sharing or have the video sites deliver contextually related ads that benefit ISP's concerns.

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:6

Re: The Article - in readable format

said by JasonOD :

So it's more of a shift from P2P to legit web video as content providers have made more available (hulu, for example), not less total traffic, just a shift.
I wouldn't be surprised to find two other factors:

First,

The "legit" content providers generally encode either less generously and perhaps also more efficiently than the "pirate" counterparts. So the overall size of the same content is smaller, even if the content providers add commercials.

Second,

But the biggest sea change is probably the declining sense that content will disappear. A lot of what I've learned about P2P has to do with psychology -- that P2P users tend to download more content today on the assumption that it will become unavailable for download tomorrow.

The "legit" sites have (finally) started to drive that fear out of the minds of viewers and I think we're only seeing the beginnings of the effect. That mindset is still very much out there, and content providers are still just getting started in getting their stuff out there.

The industry has to be careful not to abuse the viewers, now. P2P users aren't a tolerant nor trusting bunch. It's just one Sony DRM scandal away from reversing this very positive trend, and too many obtrusive distractions embedded in their content will also kill progress.
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More fun, more features, Join BroadbandReports.com, it's free...

Linklist
Premium
join:2002-03-03
Longport, NJ
kudos:5

Note that Capex costs may not be what Burstein says

Burstein's OPINION:
The current projections are they can do these upgrades without raising capex, bringing per bit costs down along a Moore's Law curve and keeping bandwidth costs per user essentially unchanged.
Karl's rewording of Burstein's OPINION:
So if the capacity costs of keeping pace with demand are nominal, does that still make AT&T's push into metered billing "inevitable?"
Please note that that wasn't the part of Burstein's post "where he quotes AT&T". I am sure AT&T would not agree that their costs aren't rising to expand the network.
--
My BLOG .. .. Internet News .. .. My Web Page
Ask yourself one question: 'Do I feel lucky?' Well, do ya punk?

funchords
Hello
Premium,MVM
join:2001-03-11
Yarmouth Port, MA
kudos:6

Re: Note that Capex costs may not be what Burstein says

said by Linklist:

Burstein's OPINION:
The current projections are they can do these upgrades without raising capex, bringing per bit costs down along a Moore's Law curve and keeping bandwidth costs per user essentially unchanged.
No, that was Dave Burstein's recounting of his AT&T source's current projections.

said by Linklist:

Karl's rewording of Burstein's OPINION:
So if the capacity costs of keeping pace with demand are nominal, does that still make AT&T's push into metered billing "inevitable?"
No, that was Karl's discussion question based on what we've just read.
--
Robb Topolski -= funchords.com =- Hillsboro, Oregon
More fun, more features, Join BroadbandReports.com, it's free...
Zoder

join:2002-04-16
Miami, FL
said by Linklist:

Please note that that wasn't the part of Burstein's post "where he quotes AT&T". I am sure AT&T would not agree that their costs aren't rising to expand the network.
If you leave out U-verse costs on the last mile, what evidence is there that AT&T's capex costs are rising to maintain and upgrade the core network? What would make telecom equipment unique from all other digital technology with regards to Moore's Law? The reports I've seen is that at&t is cutting capex. So either costs are dropping or they are cutting back on network maintenance.

scrummie02
Bentley
Premium
join:2004-04-16
Arlington, VA
Reviews:
·Verizon FiOS
·Comcast

I love Hulu

I also have a netflix account and like to stream those as well. If you have a MCE of windows you're set. I hope someone writes an MCE Hulu plugin. In any case, this is something that will get more and more popular. Broadband companies better get their infrastructure in order because VOD is the way of the future.
--
Nomads 32 Slot TF2 Server Instant Respawn - Warpath, Goldmine, Dustbowl. Come and play!
216.6.235.161

BF69
Premium
join:2004-07-28
Camden, TN

p2p fixes itself

it make sense if the media companies would makes access to content easier and less draconian then people would use their services instead of p2p. before sothparkstudios.com made episodes available for viewing how many people got those shows illegally? How many do so now? You can watch every episode ever made and you see a few 15 second/30 Second commercials per episode which isn't too annoying. small prices for getting to watch it for FREE a price most honest people are willing to pay.

Now you have Hulu so the media companies are starting to get it. Sort of. Hulu still needs more shows only has a partial seasons of some shows. Who wants to watch a partial season? At least with SouthPark Studios you can see every episode ever made.

For those of use that want to purchase shows and movies they need to come down on their prices. when you price a DVD cheaper than a download there's a problem. DVDs have manufacturing costs, delivery costs packaging costs. Downloads should be cheaper. And $2 for a TV episode is way too high even for an hour show and especially for something like Robot Chicken or Aqua Teen Hunger Force which are 12 minutes. If a Tv shows has 150 episodes you expect me to pay $300 for that? Especially when I still see it on TV? yeah lots of commercials and I can't choose which episode I see, but small price to pay compared with $300. And also an incentive for those without morals to get them off p2p.

anon2day11

@comcast.net

Answer to why AT&T noticed this

.... AT&T also saw a 50%+ reduction in total traffic (including the p2p) off their backbone because Comcast moved all their traffic off of them to their own backbone and direct peering relationships.

ptrowski
Got Helix?
Premium
join:2005-03-14
Putnam, CT
kudos:4
Reviews:
·VOIPo

Re: Answer to why AT&T noticed this

said by anon2day11 :

.... AT&T also saw a 50%+ reduction in total traffic (including the p2p) off their backbone because Comcast moved all their traffic off of them to their own backbone and direct peering relationships.
Where it can be easily "managed" for network congestion.
--
"A religious war is like children fighting over who has the strongest imaginary friend."

Have you been touched by his noodly appendage? »www.venganza.org
Kord

join:2006-10-27

Whats is there to DL?

All the new movies and music suck this year...so there is your reduction in P2P
--
I haven't failed, I've found 10,000 ways that don't work.
bluedyedvd

join:2007-04-15
Overland Park, KS

At&t was most likely correc t about p2p decline

AT&T writes off that decline in P2P use as a statistical anomaly created by a heavy mix of new customers who don't use P2P. it's alot easier to point and click on a site like youtube than it is setting up bittorent or even limewire. It does not take long though to doscover p2p

Capt Obvious

@spcsdns.net

How'd they do that?

How would an ISP be able to tell you what % of their traffic that p2p or anything else was, unless they uh, you know, used DPI?

Monday, 20-May 17:18:25 Terms of Use & Privacy | feedback | contact | Hosting by nac.net - DSL,Hosting & Co-lo
over 13.5 years online © 1999-2013 dslreports.com.