Top-class BitTorrent study published

[TankGirl]

An exceptionally good academic p2p study has just been published on The Register. The target of the study was the massive BitTorrent network operating around SuprNova, the leading torrent link site up to a few days ago when it was closed down, presumably due to the MPAA’s recent campaign against torrent and ed2k link sites. The various measurements and monitoring of the network were done about a year ago, around Christmas time. The timeframe included the p2p release of the Lord of the Rings III movie which gave the researchers an opportunity to monitor the network performance during the distribution of a highly popular file.

The scope of the study was wisely chosen, including not only the p2p traffic but the entire infrastructure involved in the release and distribution process:

Together, BitTorrent and Suprnova form a unique infrastructure that uses mirroring of the web servers with its directory structure, meta-data distribution for load balancing, a bartering technique for fair resource sharing, and a P2P moderation system to filter fake files.

The study confirmed that the reputation of BitTorrent as a fast download source is well founded: the average download speed was 240 kbit/s - more than what many broadband connections can even handle. The huge popularity of BitTorrent was also firmly established: the number of ongoing downloads varied around 400,000 - 500,000, dropping at worst under 250,000 (when both SuprNova and many trackers were experiencing difficulties). And these numbers are only for the BitTorrent client population associated with SuprNova, there were many other popular link sites available at the time.

All BitTorrent users know that torrents have a limited lifetime, but this lifetime turned out to be very hard to predict. The basic p2p availability principle applies also here: as long as there is a single complete source willing to share an item, the item remains globally available. But as the download speeds on BitTorrent depend heavily on the number of available sources the system performs much better on ‘hot’ items than on the ‘cold’ ones. Sudden rushes after much-wanted new releases are common on BitTorrent and have been coined with the term ‘flashcrowd effect’.

Here are the concluding thoughts of the researchers:

One of the big advantages of BitTorrent/Suprnova is the high level of integrity of both the content and the meta-data due to the working of its global components. We have shown that only 20 moderators combined with numerous other volunteers solve the fake-file problem on BitTorrent/Suprnova. However, this comes at a prize: system availability is hampered by the global nature of these components. Decentralization would provide an obvious solution, but makes the meta-data more vulnerable. Also, a decentralized scheme such as in Kazaa has no availability problems but lacks integrity, since Kazaa is plagued with many fake files. Clearly, decentralization is an unsolved issue that needs further research.

Another future design challenge for P2P file sharing is creating incentives to seed. For example, peers that seed files should be given preference to barter for other files.

The practical implementation of the study also reveals the extreme vulnerability and insecurity of the BitTorrent system. Using automated scripts, a donated server and a little spyware the researchers were able to track and record pretty much everything that went on on the network: all the IP numbers of seeds and downloaders, their online habits etc. Anybody else with some technical expertise could have (and may have) done the same thing. The only security measure that worked was the screening process: the researchers tried to inject a number of fake files to the network and failed to get any of them through the human moderators.

BitTorrent was not designed for a hostile environment where copyright nazis hunt down downloaders and spyware companies try to snoop what people are interested in. As the designer Bram Cohen puts it, it is rather a tool for cheap legit mass distribution of digital products:

You have a great product, many customers, and are delivering your product to hordes of happy customers online. Serving large files creates problems of scaling, flash crowds, and reliability. As you grow, they become more central to your business, but your bandwidth costs go up as well. It's a vicious cycle.

There is a solution. BitTorrent is a simple software product which addresses all of these problems.

The key to cheap file distribution is to tap the unutilized upload capacity of your customers. It's free. Their contribution grows at the same rate as their demand, creating limitless scalability for a fixed cost.

The recent MPAA campaign may well lead to the birth of a 'second generation BitTorrent' with some of the most vulnerable centralized functions being decentralized. There are limits though with how far you can go in adding security to the inherently insecure protocol.

- tg

[multi]

___


still cracking up at this slashdot comment
thanx whoever u are , made my morning...

Name the next generation P2P client something like FuckTheRIAADickhead####s. It would be interesting to see it get mentioned in the news each time RIAA sues something related to that P2P network. Call the "servers" instead "ejaculators" or something worse, and go on like that to introduce terms that violate various taboos. Soon enough, it can't get mentioned in the news anymore and (...now I get to my point, and now you will understand I'm not crazy, now you will see how this idea will triumph and free information once and for all...) RIAA's plans to scare customers by getting sue news in the newspapers won't work anymore!

HA HA HA!

Are you listening RIAA!?

We have you now!!!

THE NERDS HAVE YOU!