Cloud computing technology has gained huge attraction from research community due to its characteristics of offering the services over internet. These services are known as: software as a service (SaaS), platform as a service (PaaS), and infrastructure as a service (IaaS) and these services are offered based on the customer demand. These service are beneficial for the industries which allows industries to start the desired computation task without investing in hardware and software and when the demand is more, then resources can be increased to accomplish the task. Despite of several promising advantages of cloud computing, this paradigm faces several challenging issue which can affect the cloud computing performance such as load balancing, energy consumption, QoS and cost for resource utilization. Several approaches have been developed to address the load balancing, energy consumption and QoS related issues but cloud service utilization cost is still considered as a challenging task for cloud users where prices are fixed for utilization of cloud service unit but dynamic traffic may cause underutilization or overutilization of resources resulting in resource wastage and extra cost, respectively. In order to deal with this issue, price prediction models are developed which are based on the time-series analysis of the given data. These techniques are developed using machine learning, artificial intelligence methods and other ensemble learning. In this work, we consider different type of price prediction model and presented a comparative analysis and the obtained performance of each model is compared with different types of data traffic.
Volume 11 | Issue 5
Pages: 102-110