Hybrid Prairie Dog and Dwarf Mongoose optimization algorithm-based application placement and resource scheduling...

Area of Science:

• Computer Science
• Distributed Computing
• Artificial Intelligence

Background:

• Fog computing is crucial for delay-sensitive Internet of Things (IoT) applications.
• Resource-constrained fog devices necessitate efficient resource management for IoT application deployment.
• Optimizing Quality of Service (QoS) for diverse IoT applications in fog environments is a complex, NP-complete problem.

Purpose of the Study:

• To develop an efficient resource management strategy for fog computing environments.
• To address the challenge of deploying diverse IoT applications while meeting QoS requirements.
• To optimize energy consumption, cost, and makespan for IoT application placement.

Main Methods:

• The study introduces the Hybrid Prairie Dog and Dwarf Mongoose Optimisation Algorithm-based Resource Scheduling (HPDDMOARS) technique.
• HPDDMOARS is a weighted multi-objective mechanism for IoT application placement, optimizing energy, cost, and makespan.
• It combines the exploration capabilities of the Prairie Dog Optimization Algorithm (PDOA) with the exploitation strengths of the Dwarf Mongoose Optimization Algorithm (DMOA).

Main Results:

• Experimental validation demonstrated significant improvements compared to baseline approaches.
• Achieved a 22.18% reduction in energy consumption.
• Reduced makespan by 24.98% and lowered cost by 18.64%.

Conclusions:

• The proposed HPDDMOARS technique effectively addresses the challenges of IoT application placement in fog computing.
• It balances exploration and exploitation phases to achieve optimal resource allocation.
• The method successfully meets QoS criteria while minimizing energy, cost, and makespan.