A hierarchical control protocol for group-oriented playbacks supported by content distribution networks

Fortino, Giancarlo; Mastroianni, Carlo; Russo, Wilma

doi:10.1016/j.jnca.2008.04.001

Cited by 8 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We carried out the experimental tests in an ad hoc network simulator [31] based on discrete events. The simulator has an event scheduler, a set of processes which simulate network nodes, a topology manager and events which represent a number of operations, such as initialisation, messages, computation, etc.…”

Section: Experimental Analysismentioning

confidence: 99%

Fault tolerant decentralised -Means clustering for asynchronous large-scale networks

Fatta

Blasa

Cafiero

et al. 2013

Journal of Parallel and Distributed Computing

Self Cite

View full text Add to dashboard Cite

The K-Means algorithm for cluster analysis is one of the most influential and popular data mining methods. Its straightforward parallel formulation is well suited for distributed memory systems with reliable interconnection networks, such as massively parallel processors and clusters of workstations. However, in large-scale geographically distributed systems the straightforward parallel algorithm can be rendered useless by a single communication failure or high latency in communication paths. The lack of scalable and fault tolerant global communication and synchronisation methods in large-scale systems has hindered the adoption of the K-Means algorithm for applications in large networked systems such as wireless sensor networks, peer-to-peer systems and mobile ad hoc networks. This work proposes a fully distributed K-Means algorithm (Epidemic K-Means) which does not require global communication and is intrinsically fault tolerant. The proposed distributed K-Means algorithm provides a clustering solution which can approximate the solution of an ideal centralised algorithm over the aggregated data as closely as desired. A comparative performance analysis is carried out against the state of the art sampling methods and shows that the proposed method overcomes the limitations of the sampling-based approaches for skewed clusters distributions. The experimental analysis confirms that the proposed algorithm is very accurate and fault tolerant under unreliable network conditions (message loss and node failures) and is suitable for asynchronous networks of very large and extreme scale.

show abstract

Section: Experimental Analysismentioning

confidence: 99%

Fault tolerant decentralised -Means clustering for asynchronous large-scale networks

Fatta

Blasa

Cafiero

et al. 2013

Journal of Parallel and Distributed Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…N S is set to {2, 3, 5, 10} to consider small and large surrogate clusters. The average latency times among architecture components (T CS , T SS , and T SO ) are set according to the following model [6]:…”

Section: Simulation Parametersmentioning

confidence: 99%

Distributed architectures for surrogate clustering in CDNs

Fortino

Garro

Mascillaro

et al. 2009

Proceedings of the 4th Edition of the UPGRADE-CN Workshop on Use of P2P, GRID and Agents for the Development of Content Network

Self Cite

View full text Add to dashboard Cite

Content distribution networks (CDNs) are the most adopted solution for an efficient content delivery over the Internet. They are usually based on caching content, originally produced and stored in origin servers, into surrogate servers which are closer to final users so as to improve the average user perceived latency related to content requests. To further improve the performance of CDNs, new caching techniques should be designed which are not limited to stand-alone surrogate servers but involve coordination among a set of surrogate servers. In this paper we therefore propose the design and evaluation of several distributed architectures for clustering surrogate server: master/slave, multicast-based and peer-to-peer. An agent-oriented modeling and simulation methodology is exploited to model and evaluate the proposed architectures in significant scenarios. The results obtained from simulation show that the designed surrogate clustering architectures allow to improve performance with respect to caching techniques of conventional CDNs.

show abstract

“…According to the recent Cisco's report in 2016, the Internet video content traffic will be 82 percent of all consumer traffic by 2020 [1], which confirms the prevalence of contentcentric Internet usage. This in turn has led to the ever-increasing demands for highly scalable and efficient content distribution and delivery [2,3,4,5,6]. An accurate prediction on future content consumption is one of the important building blocks for such demand.…”

Section: Introductionmentioning

confidence: 99%