An intelligent model for supporting edge migration for virtual function chains in next generation internet of things.

Area of Science:

• Computer Science
• Distributed Systems
• Internet of Things

Background:

• Edge computing leverages IoT advancements for local data processing, addressing limitations of cloud infrastructures.
• Virtual-Function-Chains (VFCs) optimize computational power at the edge for demanding services.
• Low computational resources on IoT nodes necessitate efficient resource management strategies.

Purpose of the Study:

• To introduce an intelligent migration model capable of supporting Virtual-Function-Chains at the edge.
• To enhance the robustness and Quality of Service (QoS) of edge-based services on resource-constrained IoT devices.
• To enable dynamic adaptation of VFCs to edge network conditions and device failures.

Main Methods:

• Development of an intelligent migration model for edge computing environments.
• Implementation of auto-healing using cold migrations for Virtual Function failures.
• Integration of a QoS monitoring model to trigger live migrations for load balancing.
• Evaluation of the model's performance and comparison with existing frameworks like Kubernetes.

Main Results:

• The proposed migration model significantly increases the robustness of edge-based services on low-powered IoT devices.
• Auto-healing capabilities ensure service continuity by recovering from unexpected Virtual Function failures.
• Live migrations effectively prevent edge device overload by dynamically reallocating VFCs based on QoS.
• The model demonstrates superior reaction capabilities to edge network fluctuations compared to Kubernetes.

Conclusions:

• The intelligent migration model effectively supports Virtual-Function-Chains in edge computing, enhancing service reliability.
• The model provides a robust solution for managing demanding services on resource-limited IoT devices.
• This approach offers a promising strategy for dynamic and resilient edge service deployment.