P2P - From Internet Scourge to Savior

microbrewer writes "The MIT Technology Review has up a feature discussing the future of p2p networks. Specifically, they look at their role in content distribution, in the age of ubiquitous video services. Soon, the article asserts, the very same p2p-style networks that 'threatened' legitimate business may be the basis for most video-on-demand services." From the article: "So how could additional P2P traffic actually be a good thing for the Internet? Carnegie Mellon's Zhang points out that because peer-to-peer networks exploit both the downlink and uplink capacities of users' Internet connections, they distribute content more efficiently than centralized 'unicast' technologies. Zhang also says it should be possible to label P2P traffic so that service providers can track it and decide how much of it to allow through their networks. He and colleagues from the University of California at Berkeley have founded a startup, Rinera, to develop software that will give service providers such control."

Comments

Re:Wait..

[Reality Master 101]

Wait. I pay for these networkes, Is ME who decide anithing. And I decide with my money to have P2P in full use, and not as 2th or 3th level.

No, you aren't paying for "these networkes", you are paying for exactly what the service agreement tells you, which is probably not unlimited usage. If you want to make the decisions, then pay the big $$$ for commercial, unrestricted internet access.

Re:Yeah but

[Capt James McCarthy]

It depends upon your ISP. Speakeasy's agreement states that I can use all of my bandwidth 24/7 without any problems. A SysAdmin's ISP.

Re:Yeah but

[microbrewer]

Intra Network bandwith is not that expensive for ISPs it when they start to share data with other networks it gets expansive .

The LX Systems techology in Peer Impact that is mentioned in the MIT article uses peer clustering techniques to keep as mach data in a ISPs domain as possible and they also use geo-location techniqies so the trafic doesnt travel long distances if it doesnt have to .

P2P not 'Long Tail' friendly

[TheWoozle]

P2P is a decent way to go for popular stuff, but it's not so great when you're looking for obscure stuff.

Re:P2P not 'Long Tail' friendly

[plasmacutter]

On the contrary.. it's been the most reliable and comprehensive way for me to access out of print and obscure international titles.

Practicality and reality

[br00tus]

As far as broadcasts over the Internet done in a technically sensible way, old-timers may remember the MBONE initiative. This would have distributed broadcast video via IP Multicast. All of the "Tier One" ISPs I knew of, as well as many Tier Two ISPs had the capability to do this, the equipment in place - all they had to do was turn IP multicasting on on their Cisco routers. But management did not want to do this, because they thought it would fill their bandwidth up with video, which they didn't want. At the time, traffic shaping and billing technology was not really up to speed, people were still used to how NSFnet did things to some extent. So instead of multicasting, people did p2p, which is less efficient. After Napster began coming under legal assault, Gnutella was released with technology to specifically evade attempts to block it.

Aside from technical issues, I think decentralization, peer-to-peer and so forth is the way to go. I don't want to be the little receiver of content from the Giant Corporation with DRM, monopoly price increases and whatnot. To me it makes sense (like Mbone did) and gives me more freedom. It allows me to publish content, which Youtube and whatnot can not censor if they wish. Which is precisely why it won't happen - we don't live in some federated decentralized anarchist council structure, we live in an imperialist, capitalist society where capital is centralized in a few hands, along with the media, political power for the most part, and so on. Which is why peer to peer decentralization has been under attack since day one.

What upload bandwidth?

[PingSpike]

Around here at least, the best a consumer can get for upload bandwidth is 384kbps...without going to a T line from the phone company. If they can't handle supporting those paltry offerings, which their customers paid a non-platry sum to get...I'm going to have a hard time mustering much sympathy up for them.

P2P topology is all wrong

[Animats]

The trouble with "P2P" in its present form is that the topology is designed to evade copyright, not minimize bandwidth. Peering nodes aren't necessarily near each other. You can, and do, get situations where the same content traverses the same backbone paths multiple times. There's no end user penalty for having faraway peers, but it generates unnecessary load.

Reminds me of, many years ago, watching two coal trains passing each other in opposite directions. You don't see that kind of stupidity any more. Somewhere a trader will do a swap, rather than physically shipping the commodity around.

Netnews does this right, assuming you want a broadcast system. Netnews was designed for slow links and bandwidth minimization. As I point out occasionally, Netnews could easily handle the entire audio output of the RIAA, which is only a few gigabytes per day, using far less bandwidth than the present "P2P" systems.

What will work is ISP-level caching. AOL does this, although in a somewhat annoying fashion. In a different way, so does Akamai. We'll probably see more of that.

DD-WRT can help.

[Kadin2048]

The DD-WRT firmware for WRT-54GL routers will do this. It can de-prioritize various kinds of packets, I suspect based on header inspection. I don't know whether it's smart enough to pick up on the obfuscated Bittorrent packets used by newer versions of Azureus (which was designed to be resistant to this sort of inspection), but it will get some of it.

I'm the "unofficial sysadmin" for my house, which is shared with several other single guys, by virtue of having the router in my room, and DD-WRT makes QoS fairly simple. Things that require real-time performance like SSH, Citrix, and online games get high priority, HTTP text transfers get medium, HTTP and FTP file transfers (don't ask me how it can tell the difference between HTTP text and HTTP file transfer) get low, and P2P apps get "bulk." This doesn't prevent P2P use, in fact at night it pretty much saturates the connection, but it does fairly well at keeping them from making other services impossible to use.

The other thing to do is just go to each of the client computers and kindly insist that they turn the maximum up and down speeds on their P2P apps down to something reasonable; that can improve the situation dramatically.

Re:Legal Use of technology

[arivanov]

That is all true especially once you see what P2P is good at.

Once you have discounted the illegal uses it becomes bloody obvious that most P2P uses are nothing but halfbaked emulation of multicast done by people with poor understanding of networking. Node discovery, node promotion to hypernode, sending single request to multiple interested parties are all trivial in a multicast environment. On top of the in a multicast environment the provider can easily enforce and control QoS, administrative boundaries, seed the network with nodes, limit propagation of requests to a network region, etc.

By the way - this is especially valid as far as Generic P2P Vide On Demand and P2P voice like Skype. These are solutions looking for a problem. The problem is currently solely in the fact that idiots in big incumbent telcos in the chase of "build it big" residential access have largely ignored issues like multicast, QoS and the like. The moment they themselves start distributing content they will reconfigure their network to support that and throttle down generic P2P so it does not stand in the way. That will be the moment when the problem will disappear and this will be the moment when P2P multimedia distribution companies will begin to die.