QoS-Aware and Fault-Tolerant Replica Placement.
international conference on algorithms and architectures for parallel processing(2020)
摘要
As emerging applications become more and more distributed and decentralized, it has become a more challenging problem to design and build fault-tolerant network systems with high Quality of Service (QoS) guarantee. In this paper, an optimal replica placement problem is formulated in terms of minimizing the replica placement cost subject to both QoS and fault-tolerant constraints. Based on the generalized graph model, the optimal replica placement problem is proved to be NP-hard. The essential properties of the proposed problem are investigated and two criteria, which can quantitatively measure the replica demand of each node and the contribution of one replica to other nodes in the graph, are proposed. The criteria are employed to develop efficient heuristic algorithms. Finally the proposed algorithms are evaluated with extensive network configurations and the experimental results show that the proposed heuristic algorithms can generate solutions very close to the optimal results.
更多查看译文
关键词
Replica placement, Quality of service, Fault tolerance, Heuristic algorithm, Distributed system
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络