Experimental evaluation of peer-to-peer applications

2008

ABSTRACT In the world of Peer-to-Peer (P2P) networking different tools and protocols have been developed to make the resource sharing and the information retrieval more efficient. In researching of large-scale networked and decentralized systems simulation tools are needed to verify those new protocols since the manageability and availability of real large networks is not simple. P2PNetSim as simulation environment allows the execution of such algorithms in a parallel and distributed way, maintaining high scalability.

ICNS 2011, The Seventh …, 2011

With the advance of Peer-to-Peer solutions, research and commercial players have shown interest in enhancing local client and overall swarm performance in order to improve content distribution and user satisfaction. Protocol measurements and careful client and swarm behavior analysis are required to provide valuable information on improving performance. In this paper, we present a Peer-to-Peer testing infrastructure that enables easy deployment of complete and controlled BitTorrent swarms. The infrastructure allows a variety of realistic scenarios to be run with the ability to configure characteristics such as client type, bandwidth management, churn rate and number of connections.

2009 17th Euromicro International Conference on Parallel, Distributed and Network-based Processing, 2009

This paper introduces the PeerSE Experiment Environment (EE) including concepts for a simulation-based analysis and implementation of P2P systems. The PeerSE EE supports a comparative analysis of P2P system models to identify models fulfilling given requirements. In a next step, model components are reused for implementing laboratory and real-world P2P systems. The integrated visualization for simulation and laboratory experiment results helps developers to compare the behavior of simulation models and laboratory systems. Thus, the PeerSE EE allows a controlled transition of model components to laboratory components. The applicability of the PeerSE EE is demonstrated by exemplarily analyzing and implementing a P2P system.

2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (ccgrid 2012), 2012

Discrete Event Simulation (DES) is one of the major experimental methodologies in several scientific and engineering domains. Parallel Discrete Event Simulation (PDES) constitutes a very active research field for at least three decades, to surpass speed and size limitations. In the context of Peer-to-Peer (P2P) protocols, most studies rely on simulation. Surprisingly enough, none of the mainstream P2P discrete event simulators allows parallel simulation although the tool scalability is considered as the major quality metric by several authors. This paper revisits the classical PDES methods in the light of distributed system simulation and proposes a new parallelization design specifically suited to this context. The constraints posed on the simulator internals are presented, and an OS-inspired architecture is proposed. In addition, a new thread synchronization mechanism is introduced for efficiency despite the very fine grain parallelism inherent to the target scenarios. This new architecture was implemented into the general-purpose opensource simulation framework SimGrid. We show that the new design does not hinder the tool scalability. In fact, the sequential version of SimGrid remains orders of magnitude more scalable than state of the art simulators, while the parallel execution saves up to 33% of the execution time on Chord simulations.

Next Generation Internet Networks, 2005, 2005

Peer-to-Peer (P2P) file-sharing has become the killer application in the wired Internet and might also be highly attractive for mobile networks. In particular since UMTS operators are searching for new applications which do both: a) exploit the potential of the UMTS technology and b) motivate the user to adopt the new technology.

2009

P2P networks are scalable in serving multimedia files over Internet which is in contrast to the centralized client-server system in which the server provides all the resources to the clients. In this research, we have designed and implemented a Peer-to-Peer network simulator to efficiently serve multimedia on the Internet considering that P2P streaming applications scale very superior even in large crowd scenarios. Simulated data was used to conduct experiment from current popular multimedia serving websites such as YouTube. We have then simulated user traffic by using the P2P system to compare the performance result with the traditional Client-Server system.

The 11th IEEE International Conference on Networks, 2003. ICON2003., 2003

Peer-to-peer (P2P) protocols implemented in wireless networks and mixed scenarios would generate new applications and value added services. In order to analyze the performance of such protocols the specific topologies of current and future wireless networks should be taken into account. In our paper we analyze the performance of a Gnutella-type protocol query engine on random mesh (Random), semi-random mesh (Semi), connected stars (ConStars) and connected mesh (ConMesh) application layer network topologies. In addition we discuss and simulate for the considered topologies the effect of adaptive time-to-live (TTL) techniques. We take into account in our analysis the document replication phenomenon. Within this paper we limit ourselves to the simulation of static network topologies. The results of this study show that Semi and ConStars are the most useful topologies. Semi is the best topology for heterogeneous devices, e.g. all of the nodes are mobile phones. ConStars is useful when the network consists of devices with different capabilities, e.g. mobile phones and PC-s. If the node degrees are chosen appropriately the bandwidth needs can reasonably small making content sharing peer-to-peer applications feasible in pure wireless networks and mixed scenarios.

2000

Simulation is an efficient way to evaluate new peer-to-peer models. It requires two implicit properties: large scale and high dynamicity. In the context of our work that proposes a peer-to-peer structure based on partitioning a de Bruijn graph and its load balancing algorithms, we developed a simulator for evaluation purposes. This paper introduces a three- layer architecture of the simulator.

11th IEEE/ACM International Symposium on Modeling, Analysis and Simulation of Computer Telecommunications Systems, 2003. MASCOTS 2003., 2003

2008

Nowadays P2P applications dominate many networks. However, despite of their diffusion, the analysis of teletraffic generated by these applications in real-time environment remains an open issue. In this paper we look at analytical and simulation models, which can be used for quantifying volumes of P2P teletraffic at the network layer, to ease still existing difficulties associated with monitoring such teletraffic. We present an event-driven simulator, able to simulate P2P teletraffic both at the application and network layers in large overlay networks (we have successfully tested it in networks used by up to 500,000 peers). The simulation model was developed for studying the impact that P2P systems have on the network layer performance. We studied also the application layer dynamics by looking at time evolution of file-downloading processes and at the offered load generated in such networks. We report the results of several simulation scenarios, in which we focused on the consequences such teletraffic has on the Internet access link of an enterprise network.