The multicast policy and its relationship to replicated data placement

Author:

Wolfson Ouri1,Milo Amir1

Affiliation:

1. Columbia Univ., New York, NY

Abstract

In this paper we consider the communication complexity of maintaining the replicas of a logical data-item, in a database distributed over a computer network. We propose a new method, called the minimum spanning tree write, by which a processor in the network should multicast a write of a logical data-item, to all the processors that store replicas of the items. Then we show that the minimum spanning tree write is optimal from the communication cost point of view. We also demonstate that the method by which a write is multicast to all the replicas of a data-item affects the optimal replication scheme of the item, i.e., at which processors in the network the replicas should be located. Therefore, next we consider the problem of determining an optimal replicaiton scheme for a data item, assuming that each processor employs the minimum spanning tree write at run-time. The problem for general networks is shown NP-Complete, but we provide efficient algorithms to obtain an optimal allocation scheme for three common types of network topologies. They are completely-connected, tree, and ring networks. For these topologies, efficient algorithms are also provided for the case in which reliability considerations dictate a minimum number of replicas.

Publisher

Association for Computing Machinery (ACM)

Subject

Information Systems

Reference19 articles.

1. Allocation of copies of a file in an information network

2. Host groups

3. Distributed process groups in the V Kernel

4. Optimal file allocation in a computer network: a solution method based on the knapsack problem;CERI S.;Comput. Netw.,1982

5. Reverse path forwarding of broadcast packets

Cited by 74 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3