Global detection of live virtual machine migration based on cellular neural networks

Area of Science:

• Cloud Computing
• Artificial Intelligence
• Computer Engineering

Background:

• Cloud environments require efficient monitoring of dynamic virtual resources.
• Existing methods for virtual machine migration detection face challenges with scale and heterogeneity.

Purpose of the Study:

• To develop an advanced algorithm for detecting live virtual machine (VM) migration in cloud computing.
• To enhance the efficiency and intuitiveness of VM migration monitoring.

Main Methods:

• A novel live VM migration detection algorithm utilizing cellular neural networks (CNNs).
• Parameter optimization for CNNs using an improved particle swarm optimization algorithm with bubble sort.
• Comparative analysis against the best fit heuristic algorithm.

Main Results:

• The proposed CNN-based method provides intuitive visualization of VM migration.
• Demonstrated reduction in processing time compared to the best fit heuristic algorithm.
• The approach is suitable for parallel and very large scale integration (VLSI) implementations.

Conclusions:

• The developed algorithm offers a more efficient and intuitive solution for live VM migration detection.
• The method's adaptability to parallelism and VLSI suggests scalability for large-scale cloud monitoring.
• This advancement improves the overall performance of VM migration detection in cloud infrastructure.