Delay Optimal Schemes for Internet of Things Applications in Heterogeneous Edge Cloud Computing Networks

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Internet of Things (IoT) applications in healthcare, intelligent vehicles, and smart homes generate delay-sensitive data requiring rapid execution.
• Existing offloading and scheduling techniques in dynamic edge computing environments often underperform for delay-sensitive IoT applications.
• Software-Defined Networks (SDN) and fog computing offer edge computing paradigms to minimize end-to-end delays for these applications.

Purpose of the Study:

• To develop an optimized framework for joint task offloading and scheduling to address the performance limitations of existing methods in dynamic IoT environments.
• To minimize the hybrid delay of delay-sensitive IoT applications by formulating the problem as a combinatorial integer linear programming (CILP) model.
• To introduce a novel Joint Task Offloading and Scheduling (JTOS) framework encompassing task offloading, sequencing, scheduling, searching, and failure management.

Main Methods:

• Formulation of joint task offloading and scheduling problems as combinatorial integer linear programming (CILP).
• Development and implementation of the Joint Task Offloading and Scheduling (JTOS) framework.
• Performance evaluation through comparative analysis against existing baseline methods in dynamic environments.

Main Results:

• The proposed JTOS framework demonstrates superior performance in minimizing hybrid delay for all applications compared to existing baseline methods.
• JTOS achieves a 39% reduction in processing delay for IoT applications.
• JTOS achieves a 35% reduction in communication delay for IoT applications.

Conclusions:

• The JTOS framework effectively addresses the challenges of offloading and scheduling for delay-sensitive IoT applications in dynamic edge computing settings.
• JTOS offers significant improvements in reducing both processing and communication delays, enhancing overall application performance.
• The CILP formulation provides a robust approach to optimizing joint task offloading and scheduling decisions for IoT applications.