Towards the Scheduling of Access Requests in Cloud Storage

Abstract

In recent years, cloud computing has emerged as an enabling technology, in which virtual machine migration and dynamic resource allocation is one of the hot issues. During the migration of virtual machine, access requests from end users coming constantly. So how to schedule these requests to ensure the performance of cloud storage system is the key issue. In this paper, we focus on scheduling access requests to provide end users desirable data when virtual data nodes are migrating. In this process, we interested in the following performance in the system: the availability of data, energy consumption and cost. The contributions of this paper are: (1) analyze the open issues of scheduling in cloud storage; (2) present the environment of scheduling access requests in cloud storage and discuss metrics which affect the scheduler; (3) define the formal concepts about this problem and give the main steps of scheduling algorithm. The main innovation of this paper is fully taken into account data node's mobility in cloud storage. So we could establish the connection between resource availability and system availability. The results of this paper can be used in scheduling and resource management in cloud storage system to provide the mass storage of streaming media in cloud and its continuous access.