Multi-Objective Task Scheduling in Cloud IoT Environments Using Differential Evaluation for Big Data Processing

In Cloud IoT environments, efficient task scheduling is critical for optimizing resource utilization, reducing latency, and enhancing overall system performance. However, current methods struggle to balance the diverse demands of such environments. This paper presents the Multi-Objective Task Scheduling in Cloud IoT Environments (MOTS-CIoTE) framework, which controls K-means clustering and the Differential Evolutionary (DE) algorithm to optimize resource allocation and minimize task completion time. Our approach achieves significant improvements across key metrics: The system has a total throughput of 827 (tasks per second), a latency of 14.13 ms, a resource contention of 0.33%, an energy efficiency of 0.48 (Joules per task), and a cost-effectiveness of 3.58. The results highlight the considerable potential of MOTS-CIoTE in tackling the intricacies of Cloud IoT environments and augmenting their efficacy, hence facilitating more efficient resource allocation and boosting user satisfaction.