The multi-tier architecture style has become an industry standard in modern data centers with each tier providing certain functionality. To avoid congestion and to adhere the SLA under fluctuating workload and unpredictable failures of Mission Critical Multi-tier applications hosted in the cloud, we need a Dynamic admission control policy, such that the requests must be processed from the first tier to the last without any delay. This paper presents the least strict admission control policy, which will induce the maximal throughput, for a twotier system with parallel servers. We propose an optimization model to minimize the total number of virtual machines for computing resources in each tier by dynamically varying the mean service rate of the VMs. Some performance indicators and computational results showing the effect of model parameters are presented. This model is also applicable to priority as well as real-time based applications in Cloud based environment.Copyright c 2016 Institute of Advanced Engineering and Science.All rights reserved.Corresponding Author: C. N. Sahoo School of Computer Engineering, KIIT University Bhubaneswar-751024, India Phone: +91-9880149447 Email: nishikant.choudhury@gmail.com
INTRODUCTIONCloud computing greatly lowers the threshold for deploying and maintaining web applications since it provides infrastructure as a service (IaaS) and platform as a service (PaaS) for web applications [1]. Consequently, a number of web applications, particularly the web applications of medium and small enterprises, have been built into a cloud environment. Meanwhile, leading IT companies have established public commercial clouds. For example, Google App Engine enables enterprises to build and host web applications on the same systems that power Google applications. App Engine offers fast development and deployment; simple administration, with no need to worry about hardware, patches or backups; and effortless scalability [2]. IBM also provides cloud options [3]. Amazon Elastic Compute Cloud (Amazon EC2) is a web service that provides resizable compute capacity in the cloud. It is designed to make web-scale computing easier for developers [4]. We even can establish a private cloud with Ubuntu Enterprise Cloud to offer immediacy and elasticity in the infrastructure of web applications [5]. In summary, both of the numbers of cloud applications and providers have kept gradually increasing for a couple of years [6,7]. As a result, computing resource scheduling and performance managing have been one of the most important aspects of cloud computing [8,9]. This paper focuses on queueing-based analytical model for performance of web based applications with multi-tiered architecture. It is quite difficult to predict the traffic in web based applications. In case of Real-time or Mission-Critical applications, requests must be processed from the 1st tier to the last without any delay. If the release and processing times of requests are known, the problem for determining the processing order of requests i...