Achieving High Availability Of Resources In Teaching And Learning System In Cloud Environment

Cloud computing environment is an excellent alternative for deploying a teaching and learning system of an educational institution which is under budget shortage in order to offer learning and teaching services to both students and teachers effectively without spending much on computers and network devices. Cloud Computing is currently one of the recent technologies which is likely to have a tremendous effect on teaching and learning. The cloud provides access to rich learning tools and materials accessible from anywhere and anytime. Students and teachers can access the cloud from any internet-enabled device using any standard browser. Through a standard browser, students and teachers can access lesson content, background material, tests and subject-specific applications. By using a cloud environment, any higher institution can create a comprehensive online environment that provides the same high-quality content and resources to every student, no matter where they are in the country, or even the world. The cloud platform can also provide students with key resources such as written content, visual and audio to supplement lectures and laboratory sessions.High availability systems aim to remain available at all times, preventing service disruptions due to power outages, hardware failures, high response time, and system upgrades. All these contribute to low system availability and failures are likely to be normal rather than exceptional. To guarantee high availability in a system in cloud environment, replicating the popular system resources to nearby nodes is a best alternative, as clients will be able to access system resources with minimal response time. In our research, we propose a High Availability Assurance Replication Strategy that takes into account the following aspects:(a) Establishment of a relationship between system availability and minimal replica number that is needed to ensure high availability in cloud environment (b) Rough estimation and preservation of least replica number for any expected availability level in cloud environment.(c) Decision on which data to replicate.(d) Efficient mechanisms to place replicas to achieve high availability and load balancing.Finally, we propose a Weighted Throttled load balancing algorithm; modifying the Throttled load balancer by assigning a weight to each virtual machine (VM) in order to achieve better response time and processing time. In this proposed load balancing algorithm using the concept of weights in Throttled load balancing, VMs are assigned different amount of the available processing power of a server node to the individual application services. To these VMs of different processing powers, tasks or requests are assigned to the most powerful VM available. Our proposed algorithm is implemented with a simulator, CloudSim. It registers a significant reduction in the response time and this improves service quality and hence high availability of the web-based resources is achieved.