Choosing replica placement heuristics for wide-area systems

Karlsson, Magnus; Karamanolis, Christos

doi:10.1109/icdcs.2004.1281600

Cited by 79 publications

(63 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As already stated, most problems are NP-complete, except for some very simplified instances. Karlsson et al [7,6] compare different objective functions and several heuristics to solve these complex problems.…”

Section: Related Workmentioning

confidence: 99%

Comparison of Access Policies for Replica Placement in Tree Networks

Benoît

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, we discuss and compare several policies to place replicas in tree networks subject to server capacity. The client requests are known beforehand, while the number and location of the servers are to be determined. The standard approach in the literature is to enforce that all requests of a client be served by a single server in the tree (Single). One major contribution of this paper is to assess the impact of a new policy in which requests of a given client can be processed by multiple servers (Multiple), thus distributing the processing of requests over the platform. We characterize problem instances for which Multiple cannot be more than two times better than the optimal Single solution, if this latter exists. For such instances, we provide a procedure which builds a Single solution with a guarantee on its cost. This is a very interesting result for applications which do not accept multiple servers for a given client, since it might be more difficult to implement such a complex strategy.

show abstract

Section: Related Workmentioning

confidence: 99%

Comparison of Access Policies for Replica Placement in Tree Networks

Benoît

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Optimal solutions for specific topologies [9,10] and heuristics for more general topologies [6,13,15] have been proposed. Content placement for dynamic client demand has been studied only very recently [3,5,8,18,14]. In [3], the authors address the problem with the goal of minimizing the number of replicas deployed in a tree topology while meeting client QoS and server capacity constraints.…”

Section: Content Distribution Networkmentioning

confidence: 99%

“…Dynamic content placement heuristics for replicating content inside a cluster-based cache are proposed in [5]. In [8], the authors propose an offline integer programming model for choosing the most cost-effective heuristic for a given system, workload and performance goal. By using specific constraints, the proposed model offers lower bounds for the cost associated with a number of different heuristics.…”

Section: Content Distribution Networkmentioning

confidence: 99%

“…A heuristic for dynamic replication of Internet applications in more general topologies was recently proposed in [14]. Optimization models targeting system availability [18] and for providing lower bounds on the cost-effectiveness of different replica placement heuristics [8] have also been proposed. Nevertheless, these studies propose and evaluate solutions for wired networks.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Content Placement for Mobile Content Distribution Networks

Aioffi

Mateus

Almeida

et al. 2004

Web Content Caching and Distribution

View full text Add to dashboard Cite

Abstract. As wireless networks increase in popularity, the development of efficient content distribution techniques to meet the growing and constantly changing client demand becomes a necessity. Previous content distribution network proposals, targeting mainly wired networks, are not adequate to handle the high temporal and spatial variability in client demand caused by user mobility. This paper proposes and analyzes a wireless dynamic content distribution network framework that replicates content in response to changes in client demand in order to reduce the total traffic over the network backbone. We propose a mathematical programming model to determine the offline optimal solution as well as an online algorithm based on demand forecasting. Using a previously developed mobility simulator, we show our new online algorithm outperforms the traditional centralized and distributed static placement approaches, reducing the total traffic over the network backbone in up to 78%, compared to the best previous approach analyzed.

show abstract

“…This paper does not discuss the expression of metadata, but focuses on the first requirement and presents a metadata management system RBMMS (Replica based Distributed Metadata Management System). Utilizing the solution of data replication [5][6][7][8] and web cache [9], RBMMS presents the concept of metadata replication and metadata cache, which uses LDAP directory server to store metadata and their replicas, also a cache module is added into metadata catalog. RBMMS achieves the following goals: 1) it provides multiple metadata replicas to support high availability; and 2) replica can be added or deleted dynamically.…”

Section: Introductionmentioning

confidence: 99%

Replica Based Distributed Metadata Management in Grid Environment

Jin

Xiong

et al. 2006

Computational Science – ICCS 2006

View full text Add to dashboard Cite

Abstract. Metadata management is one of the key techniques in data grid. It is required to achieve two goals: high efficiency and availability. This paper presents a Replication Based Metadata Management System (RBMMS) as metadata server implemented in Global Distributed Storage System (GDSS). To address the above two goals RBMMS maintains a spares strongly connected graph to describe replica structure and relations among the replicas. The graph is used to propagate updating information and replica discovery in the process of replica addition and removal. Cache module is also implemented in RBMMS to further improve the performance of metadata access. The evaluation demonstrates that RBMMS attains high availability and efficiency of metadata management system.

show abstract

Choosing replica placement heuristics for wide-area systems

Cited by 79 publications

References 11 publications

Comparison of Access Policies for Replica Placement in Tree Networks

Comparison of Access Policies for Replica Placement in Tree Networks

Dynamic Content Placement for Mobile Content Distribution Networks

Replica Based Distributed Metadata Management in Grid Environment

Contact Info

Product

Resources

About