DDSS: Dynamic dedicated servers scheduling for multi priority level classes in cloud computing

Abstract

Simplicity of usage, flexibility of data access, ease of maintenance, time and energy efficiency, and pay as you go policy have increased the usage of cloud computing over traditional computing. Cloud computing should be able to meet the performance expectations of different classes of customers. On the contrary, inefficient scheduling algorithms decrease the quality of service experienced by users. To improve quality of service, there are several proposed scheduling algorithms in the literature. However, these scheduling algorithms are limited because they do not consider different types of customers. Our objective is to satisfy performance expectations of customers by proposing an efficient Dynamic Dedicated Server Scheduling (DDSS) while considering different types of customers. Results show that the customer drop rate and throughput can be significantly improved by DDSS. Our proposed scheduling and related analysis will help cloud service providers build efficient cloud computing service architectures through considering different types of priority class performances such as, drop rate, throughput, and utilization.