Using light weight container a mesh based dynamic allocation task scheduling algorithm for cloud with IoT network

• Published in: International journal of information technology (Singapore. Online) 2024, Vol.16 (5), p.2847-2861
• Main Authors: Shakya, Santosh, Tripathi, Priyanka
• Format: Article
• Language: English
• Subjects: Artificial Intelligence; Computer Imaging; Computer Science; Image Processing and Computer Vision; Machine Learning; Original Research; Pattern Recognition and Graphics; Software Engineering; Vision
• ISSN: 2511-2104; 2511-2112
• DOI: 10.1007/s41870-024-01772-7

Due to their flexible deployment, portability, and scalability, containers have emerged as the most promising lightweight virtualization solution for cloud services, particularly in microservices, smart Internet of Things (IoT), and cloud computing. Due to the different nature of the workload and cloud resources, the scheduler's component plays a crucial role in cloud container services with the virtual machine (VM) to improve performance and save costs. The selected container is often busy in pre-operation and denies accepting the new job, and then the VM dynamically chooses the next container to complete the task. Smartphone and IoT device usage has grown substantially, as have cloud computing-based mobile cloud applications. Vehicles, Augmented Reality, e-transportation, 2D/3D games, e-healthcare, and education are just a few industries that utilize these applications; current cloud-based frameworks offer these services on Virtual Machines. Because cloud providers operate on such a vast scale, even minor performance degradation concerns might result in a sharp increase in energy or resource utilization costs. By consuming less energy, cloud providers might potentially increase cost reduction. The quantity of energy used can be decreased using clever task-scheduling algorithms to distribute user-deployed workloads to servers thorough the container filled VMs. This work proposed a new container-based VM task scheduling "mesh-based dynamic allocation task scheduling" (MB-DATS) algorithm, which is very helpful for reducing the load on any server through container migration. Our findings were contrasted with those of the Grey Wolf Optimization (GWO), Whale Optimization Algorithm (WOA), priority-aware workloads scheduling algorithm (PA-CCWS), and Decision tree learning and Monte Carlo tree search (DT-MCTS) methods were assessed by taking the parameter load variation and makespan into account. We contrast the suggested approach with the current algorithms regarding makespan time, resource utilization, and overall task completion time.