Darshna Dalvadi¹ and Keyur Shah², ¹Ahmedabad University, India and ²Kadi Sarv Vidhyalaya, India

Content Delivery Networks have been widely used for many years to serve million of users. Lately, many of these networks are migrating to the cloud for its numerous advantages such as lower costs, availability and dynamism in resource provisioning for obtaining overall increased performance. This paper introduces a new approach towards load balancing as well as lower response time(limited latency) in cloud-based content delivery networks allowing for providing overall improved performance within the tradeoff between power consumption and quality of experience (QoE). New proposed adaptive server activation/ deactivation model aims towards switching off unutilized servers at the data center to reduce the power consumption and also to use available resources at maximum efficiency while maintaining its performance. Latency can be lowered by only shifting the load off a datacenter when it is almost fully loaded.

Cloud Computing, Content Delivery, Load Balancing, Power reduction, Resource Management, Adaptive Self Control, Performance Modeling.