Looking For “Handprints” Across P2P Networks

Slow peer-to-peer transfers can be brutal. The progress bar in your desktop client of choice is creeping along at 10kb/sec and you’re just sitting there wondering what’s wrong. Is it a problem on your end that’s preventing you from flying through the process? Or is it something completely out of your control?

It could be configuration issue with [insert name of P2P utility here]. Perhaps a conflict with a firewall or a discrepancy made between a terminal and a router. But there are cases in which slow download speeds are due to outside forces. So some computer scientists associated with Carnegie Mellon and Purdue universities have been at work developing a new protocol to help files transferred across peer-to-peer networks get to their respective destinations a little more quickly. A lot more quickly, even.

The SET protocol, as it’s called, was released for attendees of the 4th USENIX Symposium on Networked Systems Design & Implementation to parse through. Those responsible for developing the SET protocol also plan to distribute the new technology more widely in the future, too.

SET is much the same as the transfer methodology we’ve come to know as P2P. A network of individual locations scattered across the net sharing files, within which the more popular and more oft requested and downloaded files are moving along to and fro more so than less popular items. If you’ve tried downloading from a single remote IP/location, you know that it arrives as slow as a snail would with a bad back – over the same distance. (I’m not sure if snails actually have backs, having chosen not to bother doing the research. Forgive me.) But request a file that happens to have hundreds, even thousands of “seeders,” and your computer is typically chomping at the oncoming river of bits with gusto.

Where SET is different, though, is in the way it looks for commonalities in files rather than just titles and metadata. They can be labeled differently or they can be items with slightly different traits which still contain a great deal of information identical to that sought originally. They can even fit both descriptions. But overall, they’re the same. Most of the time.

So with the added smarts SET brings to the proverbial table, the “seeds” detected as credible matches can be much greater in number. There will certainly be exceptions. Not everyone shares everything, so users of the SET protocol will sometimes come to experience a slow download at one time or another, but it’s logical to presume that an encounter with a feeble bit rate would happen in fewer instances.

SET is no revolution of P2P technology. It’s the evolution of something that came before. But if you were given the option to choose the old (what you’re using presently) or the new (SET), which would you prefer? You’ll need to take into consideration that only legal, copyright-free downloads can be transferred. To avoid potential repercussions, of course. You wouldn’t want to make SET a vehicle for illicit activity, would you?