Re: [p2pi] Real life torrent statistics

Sorry that I missed this number in my notes .

The dataset includes 19333 unique <ip,port> pairs ( some IPs are repeatable)
from 1459 ASs. Or 13.25 peers/AS in average.

While aggregated by "ISP name", I got 1368 unique ISP names ( not very
accurate estimation by second part of AS name) or 14.1 peers per "ISP"

  _____  

From: Hao Chen [mailto:chsoft@gmail.com] 
Sent: Monday, August 18, 2008 2:43 AM
To: Stas Khirman
Cc: p2pi@ietf.org; p4pwg@yahoogroups.com
Subject: Re: [p2pi] Real life torrent statistics

How many ASes does the overlay span totally? It seems the hotspot effect is
not very obvious as compared to previous study. What about cluster effect of
overlay by which I mean the ratio of neighbors of the peer in the same AS or
ISP? 

2008/8/18 Stas Khirman <stas@khirman.com>

To estimate a feasibility of ALTO/P4P for real life torrents , I collected
<ip,port> information for peers from one of the most popular "PirateBay"
torrents ( almost 20k peers) and maped their IPs to corresponded ASs. Please
find attached my working notes with some interesting statistics.

In short - for such kind of swarm (20k peers), almost half of peers belong
to ASs with 100 or more peers, almost 2/3 of peers belong to ASs with 50 or
more peers. Less then 15% of peers belong to ASs with 10 or less
"neighbors".

US broadband ISPs (Road Runner, Comcast, AT&T) have about 2-3% of total
peers each.

Also, I find it surprising geo distribution of the peers - majority were in
UK , not in US (probably because content is available in US theaters).
Places 3-5 taken by Sweden, Poland and Canada (in total - more peers then in
US).

Certainly, observed "heavy" neighboring of peers is a function of swarm
size. I intend to investigate a few medium/small size swarms to have a
multi-point picture for any future discussions.

I would appreciate comments, corrections and additional ideas on data
collection and analysis. 

_______________________________________________
p2pi mailing list
p2pi@ietf.org
https://www.ietf.org/mailman/listinfo/p2pi

-- 
Cheers,Hao

Fudan University