Hi Sebastian, On Thursday 18 March 2010 5:42:10 pm Sebastian Kiesel wrote: > Hi, > > a technical concern regarding the provisioned bandwidth attribute: > > If a P2P app uses this information for a peer selection strategy like > "try the highest bandwidth tier peers first" or "try the highest > bandwidth tier peers with higher probability", and if there is a swarm > with many peers but very few of them are in the highest bandwidth tier, > the user experience of these premium customers might be worse than for a > customer with lower provisioned bandwidth (even if the P2P app uses flow > control and congestion control, the initial syn packets alone could > overwhelm the premium customer's links). > > So we should be very clear what we want to do with this kind of > information in the application.
I think this is a very good point. My personal opinion is that it would be cleaner to avoid third-party inquiries for provisioned bandwidth. For example, this would permit a peer to locally make some estimation (perhaps its "spare" capacity), and advertise that instead. That said, avoiding third-party inquiries also has a couple of drawbacks: 1) In order for a third-party to determine "provisioned" bandwidth, there must be an ALTO Client running at the target IP address. Then, to distribute the provisioned bandwidth to another ALTO Client, it either must be delivered directly directly or through an intermediary using an application-specific protocol (e.g., a P2P tracker, peer exchange, or a DHT). 2) An ALTO Client can "lie" about its provisioned bandwidth. Actually, this could also be seen as an advantage, since it may not be "lying" for malicious reasons. Instead, it may be performing reasonable "adjustments" based on other applications that are also sucking up some of the bandwidth. Using the ALTO redistribution mechanism (where the value is signed by the ALTO Server) would avoid the possibility of lying, but it also means that no such "adjustments" would be possible. Of course, an ALTO Client still has the option to not redistribute its own provisioned bandwidth value in the first place. -- Richard Alimi Department of Computer Science Yale University _______________________________________________ alto mailing list [email protected] https://www.ietf.org/mailman/listinfo/alto
