In recent years, container-based virtualization has gained popularity due to its ease of deployment and agility in cloud resource provisioning. The traditional virtual machine (VM) is based on modern innovation, has superseded technol- ogy in cloud computing which is known as containerization technology, and it is superior in terms of overall performance, reliability and efficiency. Con- tainerized clouds deliver superior performance because they make the most of the resources available at the host level and make use of a load-balancing strat- egy. In order to accomplish this goal, the focus of this article is on equitably dividing of the workload across all of the available servers. In this research, we proposed a Honeybee Mating Algorithm (HBMA) to combat the issue of load balancing in the container-based cloud environment by considering the deadline of tasks. We compared our findings to those of the Grey Wolf Opti- mization (GWO), Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) Algorithms. We assessed the performance of the proposed methods by considering the impact of parameters such as load variation and makespan. According to the findings of our proposed method, almost the tasks were com- pleted within the deadline, and the HBMA performed significantly better than any of the other strategies in terms of load variance and makespan.