Optimal resource allocation of dynamic video streaming applications in a public cloud environment

With the technology improvements, managing a large amount of multimedia objects such as audio, video, picture or a combination of these has become possible. Multimedia data needs more real time storage and high data transfer than traditional textual and numeric data. In addition to these requirements, significant amount of computation is demanded for multimedia applications to serve many users at the same time. Due to these reasons, cloud computing is the emerging technology necessary for the multimedia information systems as well as multimedia objects. Cloud computing provides organizations to emphasize their own business values while freeing them from setting up hardware and software infrastructures to deploy the applications. Because there are several cloud providers with different quality of service requirements, different service level agreements (SLA) and uncertainties in demand, price and availability, optimization of these issues bring out new challenges. The aim of this study is to optimize the performance and cost of a cloud service provider's storage and server systems by scheduling videos onto cloud computing environments based on SLA which specifies the performance or Quality of Service (QoS) attributes. In this paper, Linear Programming (LP) and Particle Swarm Optimization (PSO) algorithms are used to schedule video requests to cloud resources while minimizing the cost of usage of cloud systems. LP, ‘First in First out’ (FIFO), basic scheduling algorithm, and PSO algorithm are compared in terms of cost savings. The results show that proposed PSO algorithm and LP algorithm yield better results than FIFO algorithm.