The cloud has occupied major role in the business organization due to its vast advantages and benefits. As the usage of cloud services are increasing rapidly, the resource allocation in the cloud faced various problems that affected the data transmission process. Various existing methods failed handle the resource allocation due to lack of optimal solutions that affected the efficiency of the cloud environment. To overcome the existing issues of resource allocation in the cloud platform, an improved optimization technique called Global Memory-Gravitational Search Algorithm (GM-GSA) is proposed in this research. The model finds the optimal solution for the process of resource allocation that improves the performance of data transmission in the cloud computing. The performance of the developed GM-GSA algorithm is evaluated in terms of reliability, energy consumption and execution time. The model has obtained better execution time of 25, 96, 97, 98, and 150 ms. The proposed GM-GSA attained minimum energy consumption scores of 35, 55, 45, 62, and 65. The model has also gained better scores of 0.85, 0.91, 0.96, 0.91, and 0.90 for 100, 200, 300, 400, and 500 numbers of tasks compared to existing particle swarm optimization (PSO) algorithm.