On Mon, Dec 1, 2008 at 8:46 PM, Ole Tange <[EMAIL PROTECTED]> wrote:
> On Thu, Nov 27, 2008 at 6:41 AM, Soh Kam Yung <[EMAIL PROTECTED]> wrote:
>> [http://spectrum.ieee.org/dec08/7027]
>>
>> I expected this article to appear earlier
>> [http://www.lugs.org.sg/pipermail/slugnet/2008-March/018775.html] but
>> still an interesting read.
>>
>> =====
>> A Fairer, Faster Internet Protocol
>
> I tried understanding it, but I cannot see how it will guard against a
> P2P application that puts a high weight on each of its connections.

Cap. Just not on bandwidth, but of some formula based on the weight
you assigned to the connection and the bandwidth its utilizing. A very
simplified example would be (not saying this is it, it's just an
example):

f(download_size, weight) = download_size * weight

If someone were to download 100GB of something with weight 10, we
would have (ignoring unit):
f(100GB, 10) = 1000

If he were to assign weight of 0.1, we'd have
f(100GB, 0.1) = 10

The problem is how to enforce such cap, which were previously
unenforceable. The paper claims it founds a way to enforce it, which
seems plausible from my casual reading.


-- 
Chris
[EMAIL PROTECTED]
[EMAIL PROTECTED]

_______________________________________________
Slugnet mailing list
[email protected]
http://wiki.lugs.org.sg/LugsMailingListFaq
http://www.lugs.org.sg/mailman/listinfo/slugnet

Reply via email to