Multi-Factor Replica Placement Strategy for Resource Failure Extenuation in Cloud Replication Environment
Fazlina Mohd Ali, Marizuana Mat Daud, Hasimi Sallehudin, Nur Arzilawati Md Yunus, Azran Ahmad, Surya Sumarni Hussein
National University of Malaysia Universiti Putra Malaysia Universiti Tenaga Nasional Universiti Teknologi MARA System
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
Cloud replication strategies have become the evolving technology for managing the growing volume and variety of data in cloud storage. Delay in data access and data loss can affect overall system performance, which is why cloud providers often rely on replication strategies to store multiple copies of data in different locations. This approach ensures the availability of many data copies, swift responses, and fault-resilience while maintaining costs affordable for both users and providers. The inevitable main challenges in cloud replication strategies are to guarantee data is continuously accessible, whilst network consumption is reduced. To prevent such issues, an effective replication strategy needs to be established in a cloud environment to enhance overall performance. This study presented a novel Multi-factor Replica Placement Strategy (MRPS) that was able to reduce the risk of resource failure in cloud replication systems effectively. The performance of the suggested algorithm was evaluated through a comprehensive experiment using the CloudSim simulator, and its performance in choosing and replicating data files for user requests was measured. On average, MRPS achieved 16% more data availability compared to DPRS and 17% compared to RS-DCSM. Meanwhile, there is a 26% improvement in network usage by MRPS, surpassing DPRS, and a 23% improvement, surpassing RS-DCSM. These results provide strong evidence of the success of the proposed MRPS and improved performance. By surmounting the problems associated with data replication in cloud environments, this research seeks to provide valuable contributions to replica placement strategies.
逐年被引趋势
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
计算机 / AICloud Computing and Resource Management
Advanced Data Storage Technologies · Distributed systems and fault tolerance
参考文献 46
此处列出前 3 条
引用本文 3
按被引量排序,此处列出前 3 条