Improved multi-strategy secretary bird optimization for efficient IoT task scheduling in fog cloud computing.

Area of Science:

• Computer Science
• Artificial Intelligence
• Distributed Computing

Background:

• Cloud computing offers scalability for IoT but struggles with latency and bandwidth for real-time applications.
• Fog computing complements cloud by extending services to edge devices, handling lightweight tasks locally.
• Restricted fog node capabilities necessitate efficient task scheduling between fog and cloud.

Purpose of the Study:

• To present an Improved Multi-Strategy Enhanced Secretary Bird Optimization Algorithm using Reinforcement Learning (IMSESBOA+RL) for IoT task scheduling.
• To reduce data processing time and enhance Quality of Service (QoS) in fog-cloud computing environments.
• To minimize latency and energy costs while maximizing resource utilization.

Main Methods:

• Developed IMSESBOA+RL, an efficient scheduling model for scalable IoT tasks.
• Employed a multi-objective methodology based on the Secretary Bird Optimization Algorithm (SBOA) for balanced exploration and exploitation.
• Integrated Reinforcement Learning (RL) for dynamic adaptation to new workloads and optimal strategy learning.

Main Results:

• The IMSESBOA+RL approach demonstrated a 19.42% reduction in makespan.
• Execution time was reduced by 18.32% compared to baseline methods.
• The algorithm effectively managed various scalable tasks from IoT applications.

Conclusions:

• The proposed IMSESBOA+RL mechanism enhances IoT task scheduling efficiency in fog-cloud systems.
• Significant improvements in makespan and execution time validate the algorithm's effectiveness.
• The integration of SBOA and RL offers a robust solution for latency-sensitive IoT applications.