Distributed Selfish Replication

Abstract.A commonly employed abstraction for studying the object placement problem for the purpose of Internet content distribution is that of a distributed replication group. In this work the initial model of distributed replication group of Leff, Wolf, and Yu (IEEE TPDS '93) is extended to the case that individual nodes act selfishly, i.e., cater to the optimization of their individual local utilities. Our main contribution is the derivation of equilibrium object placement strategies that: (a) can guarantee improved local utilities for all nodes concurrently as compared to the corresponding local utilities under greedy local object placement; (b) do not suffer from potential mistreatment problems, inherent to centralized strategies that aim at optimizing the social utility; (c) do not require the existence of complete information at all nodes. We develop a baseline computationally efficient algorithm for obtaining the aforementioned equilibrium strategies and then extend it to improve its performance with respect to fairness. Both algorithms are realizable in practice through a distributed protocol that requires only limited exchange of information.

show abstract

“…The proof for Proposition 1 and subsequent ones can be found in a longer version of this article [12].…”

Section: A Two-step Local Search Algorithmmentioning

confidence: 99%

“…Finally, Section 8 concludes the article. The omitted proofs for the presented Propositions, as well as implementation details and more numerical results, can be found in a longer version of this article [12].…”

Section: Introductionmentioning

confidence: 99%

Local Utility Aware Content Replication

Laoutaris

Telelis

Zissimopoulos

et al. 2005

NETWORKING 2005. Networking Technologies, Services, and Protocols; Performance of Computer and Communication Networks; Mobile A

View full text Add to dashboard Cite

show abstract

“…Issues emerging from the decentralization typically concern the selfishness of involved servers (assuming that they belong to different owners, e.g. Internet Service Providers), and are considered with the use of algorithmic game theory (Chun et al 2004;Khan and Ahmad 2009;Laoutaris et al 2006). In contrast, in this paper we assume that all servers are under control of a single agent, and the goal is to design a mechanism allowing to optimize the system with minimal exchange of information between server nodes.…”

Section: Related Workmentioning

confidence: 99%

Decomposition algorithms for data placement problem based on Lagrangian relaxation and randomized rounding

Drwal

Józefczyk

2013

Ann Oper Res

View full text Add to dashboard Cite

The data placement problem arises in the design and operation of Content Delivery Networks-computer systems used to efficiently distribute Internet traffic to the users by replicating data objects (media files, applications, database queries, etc.) and caching them at multiple locations in the network. This allows not only to reduce the processing load on the server hardware, but also helps eliminating transmission network congestion. Currently all major Internet content providers entrust their offered services to such systems. In this paper we formulate the data placement problem as quadratic binary programming problem, taking into account server processing time, storage capacity and communication bandwidth. Two decomposition-based solution approaches are proposed: the Lagrangian relaxation and randomized rounding. Computational experiments are conducted in order to evaluate and compare the performance of presented algorithms.

show abstract

“…Unfortunately, the models proposed captured mainly steady-state behavior of the participants, as early P2P systems were focused on file-sharing. In file-sharing (and similarly, in selfish caching problem [12,24]), it is possible to formulate a static (not time-dependent) utility function, which can express the gain each user gets from the system in a closed formula. In the context of distributed resource management, such a solution is proposed in [34] and focuses on maintaining good relationships of a node with its neighbors by accepting neighbors' jobs to be executed on the node and therefore increasing the probability that the node's jobs will be accepted by its neighbors in the future.…”

Section: Game-theoretic Approachesmentioning

confidence: 99%

Promoting cooperation in selfish computational grids

Rządca¹,

Trystram²

2009

European Journal of Operational Research

View full text Add to dashboard Cite

In distributed computing the recent paradigm shift from centrallyowned clusters to organizationally distributed computational grids introduces a number of new challenges in resource management and scheduling. In this work, we study the problem of Selfish Load Balancing which extends the well-known Load Balancing (LB) problem to scenarios in which each processor is concerned only with the performance of its local jobs. We propose a simple mathematical model for such systems and a novel function for computing the cost of the execution of foreign jobs. Then, we use the game-theoretic framework to analyze the model in order to compute the expected result of LB performed in a grid formed by two clusters. We show that, firstly, LB is a socially-optimal strategy, and secondly, for similarly loaded clusters, it is sufficient to collaborate during longer time periods in order to make LB the dominant strategy for each cluster. However, we show that if we allow clusters to make decisions depending on their current queue length, LB will never be performed. Then, we propose a LB algorithm which balances the load more equitably, even in the presence of overloaded clusters. Our algorithms do not use any external forms of compensation (such as money). The load is balanced 1 only by considering the parameters of execution of jobs. This analysis is assessed experimentally by simulation, involving scenarios with multiple clusters and heterogeneous load.

show abstract

Distributed Selfish Replication

Cited by 107 publications

References 26 publications

Local Utility Aware Content Replication

Local Utility Aware Content Replication

Decomposition algorithms for data placement problem based on Lagrangian relaxation and randomized rounding

Promoting cooperation in selfish computational grids

Contact Info

Product

Resources

About