Design of a Machine Learning-Based Intelligent Middleware Platform for a Heterogeneous Private Edge Cloud System.

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Smart city and 5G applications require low latency, high data rates, and substantial computing resources.
• Existing heterogeneous private edge cloud systems are complex and dynamic, posing management challenges.
• Efficient infrastructure utilization is key to enhancing performance and mitigating risks.

Purpose of the Study:

• To propose an intelligent middleware platform for efficient management of heterogeneous private edge cloud infrastructure.
• To support diverse smart city and 5G network applications with varying resource demands.
• To optimize resource allocation for improved network and application performance.

Main Methods:

• Development of a novel intelligent middleware platform.
• Leveraging regression analysis and reinforcement learning for resource allocation.
• Implementation of parallel transmission, dynamic interface allocation, and machine learning algorithms.
• Utilization of container and device virtualization for hardware and environment compatibility.

Main Results:

• Demonstrated efficient allocation of heterogeneous resources to application tasks.
• Improved network and application performance through advanced techniques.
• Successfully addressed challenges of heterogeneous hardware and execution environments.
• Enhanced data rates, reduced latency, and minimized data privacy risks.

Conclusions:

• The proposed intelligent middleware platform effectively manages complex heterogeneous private edge cloud systems.
• The platform enhances the performance and reliability of smart city and 5G applications.
• This approach offers a scalable solution for future network and computing infrastructures.