当前位置: X-MOL 学术Inf. Syst. E-Bus. Manage. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A novel dynamic data replication strategy to improve access efficiency of cloud storage
Information Systems and E-Business Management ( IF 2.775 ) Pub Date : 2019-08-06 , DOI: 10.1007/s10257-019-00422-x
Sujaudeen Nannai John , T. T. Mirnalinee

Cloud computing provides on demand services to cloud users, and one among them is storage. Currently, large amount of data gets generated and demand an enormous storage. Users can avail the privilege to store their data remotely and can access them through Internet. Of course, the adoption of cloud lends the kind of storage that the user wants. Since data gets accumulated, the time it takes to store and retrieve the data is very long and difficult. Also, unfortunately the existing method of storage is to be optimized for better performance. The factors that affect the performance of cloud storage are response time, data availability and migration cost. Hence to improve these factors the data can be replicated to multiple locations. The decision on which data to be replicated, number of replicas to be created, where the replica has to be placed, management of the replicated data and the provision of optimal replica to the user are the major challenges involved in dynamic replication. We intend to propose, a novel dynamic data replication strategy with intelligent water drop (IWD) algorithm to address the challenges of replication and for the management of cloud storage. The popularity and size of the data are considered for replication. A swarm intelligence based optimization algorithm named IWD algorithm is used to optimize the process of replication and management of storage in cloud. We have compared our D2R-IWD algorithm with popular optimization techniques such as PSO, GA and found out that our methodology gives better result in terms of access efficiency for several test cases thereby improve the performance of cloud.

中文翻译:

一种新颖的动态数据复制策略,可提高云存储的访问效率

云计算为云用户提供按需服务,其中之一是存储。当前,生成大量数据并需要巨大的存储空间。用户可以利用特权来远程存储他们的数据,并可以通过Internet访问它们。当然,采用云可以提供用户所需的存储类型。由于积累了数据,因此存储和检索数据所花费的时间非常长且困难。同样,不幸的是,现有存储方法将被优化以获得更好的性能。影响云存储性能的因素是响应时间,数据可用性和迁移成本。因此,为了改善这些因素,可以将数据复制到多个位置。决定要复制哪些数据,要创建的副本数,副本必须放置的位置,复制数据的管理以及向用户提供最佳副本是动态复制所涉及的主要挑战。我们打算提出一种具有智能水滴(IWD)算法的新型动态数据复制策略,以解决复制的挑战并管理云存储。数据的受欢迎程度和大小将考虑进行复制。一种基于群体智能的优化算法,称为IWD算法,用于优化云中存储的复制和管理过程。我们将D2R-IWD算法与流行的优化技术(例如PSO,GA)进行了比较,发现我们的方法在几个测试用例的访问效率方面都给出了更好的结果,从而提高了云的性能。
更新日期:2019-08-06
down
wechat
bug