المستخلص: |
Recently, cloud computing become a new global trend of computing. It is a modern style of using the power of Internet and wide area network (WAN) to offer resources remotely. It’s a new solution and strategy to achieve high availability, flexibility, cost reduced and on demand scalability. However cloud computing has many challenges such as poor resource utilization which has deep impact in the performance of cloud computing. These problems arisen due to the huge amounts of information. So the need for efficient and powerful cloud computing load balancing algorithms is one of the most important issues in this area to improve the performance of cloud computing. Many researchers proposed various load balancing and job scheduling algorithms in cloud computing, but there is still some inefficiency in the system performance and load still imbalance. Therefore, in this research we propose a load balancing algorithm to improve the performance and efficiency in heterogeneous cloud computing environment. We propose a hybrid algorithm based on randomization and greedy algorithm, it takes advantages of both random and greedy algorithms. The algorithm considers the current resource information and the CPU capacity factor to achieve the objectives. The hybrid algorithm has been evaluated and compared with other algorithms using Cloud Analyst simulator. The results showed improvements on average response time and on processing time by considering the current resource information and the CPU capacity factor compared with other algorithms, and this means the performance has improved.
|