Tuijin Jishu/Journal of Propulsion Technology ISSN: 1001-4055 Vol. 44 No. 3 (2023) __________________________________________________________________________________ 2149 IOT (Internet Of Thinks) Resource Scheduling Problem Based on Two Imperialist Competitive and Genetic Algorithms Soheil Shakibaee Department of Mechanics, University of Aveiro, Aveiro, Portugal, Abstract:-The Internet of Things (IoT) is an emerging concept in the world of Information Technologies and Communications. Using the IoT in the cloud computing platform is also feasible. Appropriate use of resources such as processors is one of the challenges of the IoT in the cloud computing platform. Hence, task scheduling and IoT resources are key issues. The IoT resource scheduling is the same as the selection of an appropriate resource to equally distribute loads in processors and maximize the efficiency of resources. This study proposes a method by which the Imperialist Competitive and Genetic algorithms are combined to resolve this problem. For this, several simulations are performed on the proposed method. Simulation results indicate that the proposed method can solve this scheduling problem at an appropriate time and computational load, which could reduce energy consumption and increase the efficiency of IoT resources in a cloud computing platform. Keywords: IoT, resource scheduling, cloud computing, imperialist competitive algorithm, genetic algorithm Introduction Cloud computing is a computational model founded on computer networks such as the Internet and provides a new pattern to provide, consume, and deliver computing services via using networks. These services include infrastructure, software, platform, and other computing resources. Cloud computing is composed of computing and cloud. By cloud, it is meant a network or a network of broad networks such as the Internet in which the user is not fully aware of what occurs there. The working flow of cloud data centers is heterogenous due to customers of various goals and uses; thus, in a complex system of this kind, complicated scalable scheduling to cover customer needs should be expected (Zhang et al., 2015). Since the problem under consideration falls under NP-Hard problems, cloud service providers cannot provide an online decision to solve the task scheduling problem in traditional ways and at an acceptable decision time. Thus, modern heuristic methods should be utilized. There are various solutions for solving scheduling problems (Chen et al., 2012 & Zhang et al., 2015). Zhang et al. (2015) presented a scalable method for problem-solving. Although scalable, this method requires a sorting process to increase the time complexity of assigning tasks to physical machines. In a study, Birkhoff (1946) used the heuristic BvN method to assign scheduling policies. To use such methods, it is required to understand the distribution of input processes. A scheduler should be able to maintain queue stability by maximizing the waiting time for any scheduling in a BvN decomposition matrix. A problem of this kind can be highly voluminous and increase time complexity. Also, many scheduling problems have been provided to increase gains for cloud service providers or to improve social network performance. Conley et al. (2015) studied the properties of the working flow of public clouds. Walker (2008) compared Amazons’ EC2 and several high-performance computing (HPC) cluster systems, while Mehrotra et al. (2012)