Cloud computing is an emerging technology and new trend for computing based on the internet. In cloud computing, the dynamic resource allocation is an important process used for the purpose of effective distribution of loads among virtual machines. The shared use of resources by the consumers without any strategy brings a range of issues and challenges in the cloud environment such as scalability, fault tolerance, reliability, availability, and energy efficiency. Utilizing dynamic resource allocation for load balancing is considered as an important optimization process of task scheduling in cloud computing. Load balancing strives to balance the workload across virtual machines to achieve optimal machine utilization. An inefficient resource allocation strategy and load balancer may overload some virtual machines while other virtual machines are idle. Therefore, In order to achieve maximum resource efficiency and scalability, exploring efficient methods and techniques, as well as the development of novel algorithms, are highly desired. Meta-heuristic optimization techniques have had an exceptional growth over the last two decades. The remarkable ability of meta-heuristic techniques is motivated scientists from different fields to solve different problems. Furthermore, such techniques can often find optimal solutions with less computational effort than optimization algorithms, iterative methods, or simple heuristics. Accordingly, this thesis proposes a new meta-heuristic load balancing algorithm with a combination of two relatively new optimization algorithms. This algorithm can well contribute in maximizing the throughput in cloud computing using well-balanced load across virtual machines. moreover, the algorithm is able to overcome the problem of entrapping into local optimum. To evaluate the performance of the proposed algorithm, the algorithm is benchmarked on eleven test functions and a comparative study is conducted to verify the results with other existing algorithms. Also, a simulation experiment is conducted to evaluate the effectiveness of the algorithm in resource allocation problem. In the simulation, we have used uniformly and normal distribution workloads in two homogeneous and heterogeneous cloud environments. The results show the proposed algorithm is effective and outperforms than other existing algorithms. Also, our proposed algorithm illustrates that there is a significant improvement in cost of energy consumption and load balancing.