Re: New version of p2prg-mythbustering

Hi,

I would suggest to add the following paragraphs in the next version of
the document.
Tell me if you believe it makes sense.

section 3.1 (Reduced cross-domain traffic)

4. Le Blond et al. [LeBlond] show using simulations based on the real AS
distribution of 6M
    of BitTorrent peers representing 214k torrents spread among 9k ASes
that the maximum inter-AS
    traffic reduction is 56%. This maximum traffic reduction is achieved
by assuming that only copy
    of each content cross each inter-AS links. It is not possible to
achieve a higher traffic reduction,
    and such a reduction will be at the cost of a significant performance
    penalty for peers. However, they show that a reduction of 40% in
inter-AS links is still achievable without
    any significant performance penalty.


section 4.1 (Increased Application Performance)

6. Le Blond et al. [LeBlond] show using controlled experiments that the
measured application performance
    is a function of the degree of congestion on links the locality
policy tries to reduce the traffic on.

Regards,
Arnaud.


Enrico Marocco a écrit :
--
Arnaud Legout, Ph.D.

INRIA Sophia Antipolis - Planète  Phone : 00.33.4.92.38.78.15
2004 route des lucioles - BP 93   Fax   : 00.33.4.92.38.79.78
06902 Sophia Antipolis CEDEX      E-mail: arnaud.legout@...
FRANCE                            Web   : http://www-sop.inria.fr/planete/Arnaud.Legout/index.html

_______________________________________________
p2prg mailing list
p2prg@...
http://www.irtf.org/mailman/listinfo/p2prg