Cloud computing is a standard way of hosting software applications and services that is booming day by day thanks to the different utilities offered to users according to their needs and contracts. Most of these services are in the form of tasks and their execution in such environments requires efficient scheduling strategies that take into account both algorithmic and architectural features. The objective is to orchestrate the suitable assignment of the submitted tasks to the available resources on the basis of various functional requirements of end users. To overcome the scheduling issue, which is an NP-hard problem, various metaheuristic algorithms are used in literature to achieve near optimal solution. For this purpose, this paper aims to perform a comparative investigation of three common metaheuristic algorithms in the optimization process such as Shuffled Frog Leaping Algorithm (SFLA), Flower Pollination Algorithm (FPA) and Gray Wolf Optimization (GWO). Both standard and synthetic workloads are employed to analyze the performance of these algorithms by evaluating its objective function in term of two metrics which are makespan and resource utilization rate. The simulation results obtained using the CloudSim framework are very satisfactory and clearly show the value of our study.