16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008) 2008
DOI: 10.1109/pdp.2008.62
|View full text |Cite
|
Sign up to set email alerts
|

Reflecting P2P User Behaviour Models in a Simulation Environment

Abstract: Abstract-In this paper, we detail our experiences with the implementation of a popular P2P file sharing system in a simulation framework. We discuss some of the salient features of our implementation, and compare packet level with application level simulators. We then explain how we reflect user behaviour models in our simulation framework. After describing the use of visualization as an important analysis tool, we finally make some observations from our experiences with P2P simulations.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
6
0

Year Published

2009
2009
2014
2014

Publication Types

Select...
3
3
1

Relationship

0
7

Authors

Journals

citations
Cited by 9 publications
(6 citation statements)
references
References 13 publications
0
6
0
Order By: Relevance
“…The best improvement seen is that in Experiment 3. By considering that the shifted recovery time is 1 Even though it is rejected, this distribution is still much closer to the empirical data than the exponential distribution. hypo-exponentially distributed with s + 1 phases and parameters sα, (s − 1)α, .…”
Section: Methodsmentioning
confidence: 68%
See 1 more Smart Citation
“…The best improvement seen is that in Experiment 3. By considering that the shifted recovery time is 1 Even though it is rejected, this distribution is still much closer to the empirical data than the exponential distribution. hypo-exponentially distributed with s + 1 phases and parameters sα, (s − 1)α, .…”
Section: Methodsmentioning
confidence: 68%
“…Aggarwal et al [1] implemented Gnutella file sharing protocol in SSFNet simulator and compared also the packet-level and the flowlevel simulation results. They reflected the user behavior models in their simulation framework.…”
Section: Introduction and Related Workmentioning
confidence: 99%
“…Furthermore, we adopt a Poisson distribution for both the session lengths and the user arrivals. We set the average session length to 40 hours, according to what reported in [13]. Furthermore, we set the average arrival rate to 1 user per second, which coupled with the above session length results in an average overlay population of 144,000 nodes in the steady state.…”
Section: B Distributed Caching Systemmentioning
confidence: 99%
“…Furthermore, we set the average arrival rate to 1 user per second, which coupled with the above session length results in an average overlay population of 144,000 nodes in the steady state. The number of resources shared by users is generated by adopting a Weibull distribution with scale = 42 and shape = 0.5, as depicted in [13]. The selection of the resources to share or request is modeled with a Zipf-Mandelbrot distribution, as proposed by in [14].…”
Section: B Distributed Caching Systemmentioning
confidence: 99%
“…In a simulation environment, multiple parameters, which can effect on system performance, can be tuned. Thus, simulation can let us perform experiments with reasonably complex system models and fairly large topologies [3]. We conduct simulation experiments to investigate the performance of our traceable P2P record exchange framework.…”
Section: Experimental Settings and Simulation Environmentsmentioning
confidence: 99%