Indonesian Journal of Electrical Engineering and Computer Science Vol. 32, No. 1, October 2023, pp. 318~327 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v32.i1.pp318-327  318 Journal homepage: http://ijeecs.iaescore.com Resource provisioning model for executing realistic workload in heterogenous internet of things environment Naveen Kumar Chowdaiah 1,2 , Annapurna Dammur 1 1 Department of Computer Science and Engineering, PESIT-Bangalore South Campus, Bengaluru, India 2 Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru, India Article Info ABSTRACT Article history: Received Nov 16, 2022 Revised Jun 16, 2023 Accepted Jun 19, 2023 Resource provisioning considering scientific or realistic workload in a heterogeneous internet of things (IoT) environment presents significant challenges in terms of execution time and energy consumption. These challenges arise due to the dynamic nature of scientific or realistic-time workloads and the diverse characteristics of IoT devices. In this study, we propose a resource provisioning model that takes into account the dynamic and real-time nature of IoT workloads in a heterogeneous environment. The model aims to allocate computational resources effectively, considering the real-time demands of IoT applications while optimizing execution time and energy consumption. Three scientific workloads have been used to evaluate the proposed model. The results have been compared with the existing models. The results show that the proposed model attains better performance in terms of reducing time and energy consumption for the execution of workload tasks. Keywords: Energy consumption Execution time Heterogenous IoT environment Resource provisioning Scientific workloads This is an open access article under the CC BY-SA license. Corresponding Author: Naveen Kumar Chowdaiah Department of Computer Science and Engineering, PESIT-Bangalore South Campus Bengaluru, Karnataka, India Email: naveenphd872@gmail.com 1. INTRODUCTION A heterogeneous internet of things (IoT) environment refers to a system where diverse devices, technologies, and platforms come together to create interconnected networks and enable a wide range of IoT applications. In this environment, a variety of IoT devices with different capabilities, such as sensors, actuators, and controllers, coexist [1]. These devices can vary in terms of computational power, memory capacity, communication protocols, and energy constraints [2]. The communication infrastructure supporting the heterogeneous IoT environment includes a mix of technologies such as Wi-Fi, Bluetooth, Zigbee, and cellular networks, allowing devices to connect and communicate seamlessly [3]. The heterogeneity in device types and communication technologies poses both challenges and opportunities in terms of interoperability, data management, and system integration [4]. However, it also enables the deployment of IoT solutions that cater to specific use cases and application requirements, leveraging the strengths of different devices and technologies [5]. Resource provisioning in a heterogeneous IoT environment is the process of efficiently allocating and managing computational resources to support the diverse devices and applications within the system [6]. With a wide range of devices, each having unique capabilities and requirements, resource provisioning becomes crucial for ensuring optimal performance and utilization. This involves assessing the capabilities of IoT devices in terms of their computational power, memory capacity, energy constraints, and communication capabilities [7]. Based on this assessment, resources can be allocated accordingly to meet the specific needs of