2011
DOI: 10.1016/j.comnet.2010.11.011
|View full text |Cite
|
Sign up to set email alerts
|

Characterization of BitTorrent swarms and their distribution in the Internet

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
36
0
2

Year Published

2011
2011
2019
2019

Publication Types

Select...
5
4

Relationship

3
6

Authors

Journals

citations
Cited by 49 publications
(38 citation statements)
references
References 15 publications
0
36
0
2
Order By: Relevance
“…wherein n x E x [η]γ represents the capacity of the total available peers at state (x) [31] [32]. And N q refers to the total demand for the content object in the network.…”
Section: A Stochastic Analytical Modelmentioning
confidence: 99%
“…wherein n x E x [η]γ represents the capacity of the total available peers at state (x) [31] [32]. And N q refers to the total demand for the content object in the network.…”
Section: A Stochastic Analytical Modelmentioning
confidence: 99%
“…Table II shows the relative arrival rates for the four scenarios. The ratio λ s λ is related to the size of swarm s compared to all swarms, while λ 2,s λ 1,s is related to the share of local peers in swarm s. In [43], the authors measured the top-AS fraction of different swarms, defined as the maximum number of peers in one AS of the swarm normalized by the size of the swarm (max i z i ∑ j z j ). The top-AS fraction was found to vary from a minimum of slightly less than 0.1, for swarms sharing international content, to a maximum of 0.5, for swarms sharing regional content.…”
Section: B Stationary Arrival Processmentioning
confidence: 99%
“…Compared to unif.,1:1+1:10, in this scenario the symmetric swarms, though more popular in ISP 1, are less popular in total than the asymmetric ones. The use of Zipf's law for the arrival intensities in scenario zipf,1:10 is motivated by recent measurements that show that the distribution of the number of concurrent peers over swarms exhibits Zipf like characteristics over a wide range of swarm sizes [44], [45]. Symmetric and asymmetric swarms are motivated by measurements that show the difference in terms of the spatial distribution of peers between contents of regional and of global interest (e.g., the popularity of movies depending on the language [45]).…”
Section: B Stationary Arrival Processmentioning
confidence: 99%
“…These scenarios are organized with different parameters and functionalities as shown in Table 1. The choice of each parameter value is conditioned to a general media-streaming scenario and is based on several works of the literature such as [7,27,[29][30][31][32]. Regarding the scenario provision capacity, there are three categories [7], which are defined as it follows.…”
Section: Scenario Characterization and Workloadsmentioning
confidence: 99%