[bittorrent] Choking based on network closeness?
c99ang at cs.umu.se
Tue Jul 12 21:05:35 EDT 2005
Brian Dessent wrote:
> Justin Cormack wrote:
>>It came up on the list before, and there were some uses for it. The
>>main problem is that no one seems to have enough information to
>>measure closeness, unless you actually work for a large ISP or run BGP.
>>If some metric that could be made available to more developers could
>>be found then some progress might be made...
> And even then it's worthless. Suppose peer A is right down the street
> from me on the same ISP but has his upload throttled very low, and peer
> B is thousands of miles away (lets say, Sweden, for the sake of
> argument) with a fat pipe and generous limits. In terms of what's best
> for the host, it should choke peer A and reciprocate as much as possible
> with B. And the protocol already does this. Any attempt to add some
> kind of "hops away" or "geographical distance" metric is just going to
> make things worse. Just say no...
Of course depending on the definition of "best". If the host have to pay
for the bandwidth to peer B, but can talk for free with peer A.
Depending on the price, the person paying the bill may prefer a slower
download over a big bill.
I've been considering experimenting with a simple ip-filter, that
classifies peers as inside or outside the ISP (and peering ISP's), i.e.
free and non-free peers. Then one could just limit upload and download
to non-free peers, and have no limits on free peers.
More information about the BitTorrent