Harini Kota
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Harini Kota has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Ganesh Reddy Karri
School of Computer Science and Engineering, VIT-AP university, Amaravati, India
1 shared publication - Gayathri Tippani
Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal 560064, India
1 shared publication - Pillareddy Vamsheedhar Reddy
Department of CSE (AIML), Keshav Memorial Engineering College, Hyderabad 500088, India
1 shared publication - Sudheer Mangalampalli
Department of Computer Science and Engineering, Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal, Karnataka, India
1 shared publication
Latest Publications
- Open Access
Priority-Aware Multi-Objective Task Scheduling in Fog Computing Using Simulated Annealing
The number of IoT devices has been increasing at a rapid rate, and the advent of information-intensive Internet of Multimedia Things (IoMT) applications has placed serious challenges on computing infrastructure, especially for latency, energy efficiency, and responsiveness to tasks. The legacy cloud-centric approach cannot meet such requirements because it suffers from local latency and central resource allocation. To overcome such limitations, fog computing proposes a decentralized model by reducing latency and bringing computation closer to data sources. However, effective scheduling of tasks within heterogeneous and resource-limited fog environments is still an NP-hard problem, especially in multi-criteria optimization and priority-sensitive situations. This research work proposes a new simulated annealing (SA)-based task scheduling framework to perform multi-objective optimization for fog computing environments. The proposed model minimizes makespan, energy consumption, and execution cost, and integrates a priority-aware penalty function to provide high responsiveness to high-priority tasks. The SA algorithm searches the scheduling solution space by accepting potentially sub-optimal configurations during the initial iterations and further improving towards optimality as the temperature decreases. Experimental analyses on benchmark datasets obtained from Google Cloud Job Workloads demonstrate that the proposed approach outperforms ACO, PSO, I-FASC and M2MPA approaches in terms of makespan, energy consumption, execution cost, and reliability at all task volume scales. These results confirm the proposed SA-based scheduler as a scalable and effective solution for smart task scheduling within fog-enabled IoT infrastructures.
Open AccessPriority-Aware Multi-Objective Task Scheduling in Fog Computing Using Simulated Annealing
The number of IoT devices has been increasing at a rapid rate, and the advent of information-intensive Internet of Multimedia Things (IoMT) applications has placed serious challenges on computing infrastructure, especially for latency, energy efficiency, and responsiveness to tasks. The legacy cloud-centric approach cannot meet such requirements because it suffers from local latency and central resource allocation. To overcome such limitations, fog computing proposes a decentralized model by reducing latency and bringing computation closer to data sources. However, effective scheduling of tasks within heterogeneous and resource-limited fog environments is still an NP-hard problem, especially in multi-criteria optimization and priority-sensitive situations. This research work proposes a new simulated annealing (SA)-based task scheduling framework to perform multi-objective optimization for fog computing environments. The proposed model minimizes makespan, energy consumption, and execution cost, and integrates a priority-aware penalty function to provide high responsiveness to high-priority tasks. The SA algorithm searches the scheduling solution space by accepting potentially sub-optimal configurations during the initial iterations and further improving towards optimality as the temperature decreases. Experimental analyses on benchmark datasets obtained from Google Cloud Job Workloads demonstrate that the proposed approach outperforms ACO, PSO, I-FASC and M2MPA approaches in terms of makespan, energy consumption, execution cost, and reliability at all task volume scales. These results confirm the proposed SA-based scheduler as a scalable and effective solution for smart task scheduling within fog-enabled IoT infrastructures.