2024 : 4 : 14
Sadoon Azizi

Sadoon Azizi

Academic rank: Associate Professor
ORCID:
Education: PhD.
ScopusId: 65456
Faculty: Faculty of Engineering
Address: Room No. 206, Department of Computer Engineering and Information Technology , Faculty of Engineering , University of Kurdistan, Sanandaj, Iran.
Phone:

Research

Title
Deadline-aware and energy-efficient IoT task scheduling in fog computing systems: A semi-greedy approach
Type
JournalPaper
Keywords
Internet of Things, Fog computing, Cloud computing, Task scheduling, Semi-greedy algorithm, Deadline-aware, Energy consumption
Year
2022
Journal JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
DOI
Researchers Sadoon Azizi ، Mohammad Shojafar ، Jemal Abawajy ، Rajkumar Buyya

Abstract

With the rapid advancement of Internet of Things (IoT) devices, a variety of IoT applications that require a real-time response and low latency have emerged. Fog computing has become a viable platform for processing emerging IoT applications. However, fog computing devices tend to be highly distributed, dynamic, and resource-constrained, so deploying fog computing resources effectively for executing heterogeneous and delay-sensitive IoT tasks is a fundamental challenge. In this paper, we mathematically formulate the task scheduling problem to minimize the total energy consumption of fog nodes (FNs) while meeting the quality of service (QoS) requirements of IoT tasks. We also consider the minimization of the deadline violation time in our model. Next, we propose two semi-greedy based algorithms, namely priority-aware semi-greedy (PSG) and PSG with multistart procedure (PSG-M), to efficiently map IoT tasks to FNs. We evaluate the performance of the proposed task scheduling approaches with respect to the percentage of IoT tasks that meet their deadline requirement, total energy consumption, total deadline violation time, and the system’s makespan. Compared with existing algorithms, the experiment results confirm that the proposed algorithms improve the percentage of tasks meeting their deadline requirement up to 1.35x and decrease the total deadline violation time up to 97.6% compared to the second-best results, respectively, while the energy consumption of fog resources and makespan of the system are optimized.