A background-based new scheduling approach for scheduling the IoT network task with data storage in cloud environment

• Published in: Cluster computing 2024-09, Vol.27 (6), p.8577-8594
• Main Authors: Shakya, Santosh, Tripathi, Priyanka
• Format: Article
• Language: English
• Subjects: Algorithms; Bandwidths; Cloud computing; Computer Communication Networks; Computer Science; Data compression; Data storage; Energy consumption; Industrial applications; Internet of Things; Operating Systems; Performance measurement; Processor Architectures; Quality of service; Resource scheduling; Resource utilization; Scheduling; Task scheduling; Virtual environments; Workloads
• ISSN: 1386-7857; 1573-7543
• DOI: 10.1007/s10586-024-04400-y

Cloud computing is very popular due to its unique features, such as scalability, flexibility, on-demand service, and security. A competent task scheduler is necessary to boost the efficiency of a cloud system, which executes several tasks at once. With the incorporation of cloud computing, the Internet of Things (IoT) has seen tremendous improvements recently. IoT devices produce various data sets they want to store in particular places on the cloud. Data and resources may be spread across several locations and accessible from the VM through suitable technology. Cloud computing has changed how resources are used, stored, and shared for industrial applications, including data, services, and applications. Virtual Machine Placement (VMP)'s main goal is to map Virtual Machines (VMs) to Physical Machines (PMs), allowing the VMs to be used to their fullest potential without interfering with the PMs that are currently running. VM performs the selected task and stores the data in a cloud location. Recently, many algorithms have worked on only VM selection or VM migration; instead of this, we also work on the data storage related to a particular IoT device in the cloud. It considerably lowers energy usage and offers a list of live VM migrations that must be completed to obtain the best solution and store the data on the cloud in a linked location. The authors present a novel scheduling technique that outperforms other widely recognized scheduling algorithms regarding load balancing in compression, specifically the quality of service parameters. In this research, we propose a Background-based task scheduling (BBTS) algorithm to decrease energy usage and boost throughput while choosing a light VM for any activity. Furthermore, the proposed approach is compared to various task scheduling methodologies, considering multiple performance metrics such as makespan time, resource utilization, success rate, and computation time. The evaluation is conducted on a task set ranging from 100 to 1000, with makespan time ranging from 55 to 654, resource utilization ranging from 35.45 to 42.13, success rate ranging from 88.63 to 96.48, and computation time ranging from 39.12 to 529.46. These metrics are compared to the corresponding algorithms (IWC, WOA, and ALO) utilized in this research study.