Many-Objective Quantum-Inspired Particle Swarm Optimization Algorithm for Placement of Virtual Machines in Smart...

Area of Science:

• Cloud Computing
• Artificial Intelligence
• Optimization Algorithms

Background:

• Particle Swarm Optimization (PSO) is a metaheuristic for finding Pareto-optimal solutions.
• Deep learning models in virtual machines require significant computational resources.
• Efficient resource allocation is crucial for cloud computing performance.

Purpose of the Study:

• To enhance the Particle Swarm Optimization (PSO) algorithm with quantum gates.
• To ensure diversity in particle populations for efficient alternative solutions.
• To optimize resource assignment for deep learning virtual machines in cloud environments.

Main Methods:

• An extended PSO algorithm incorporating quantum gates was developed.
• The algorithm was applied to a multi-criteria optimization problem for virtual machine resource allocation.
• Simulations were conducted on an OpenStack-based laboratory cloud platform.

Main Results:

• The quantum-inspired PSO algorithm demonstrated improved solution diversity.
• The proposed method effectively addressed seven criteria including power, reliability, and cost.
• Numerical experiments confirmed the superiority of the multi-objective quantum-inspired PSO over other metaheuristics.

Conclusions:

• Quantum gates enhance PSO by ensuring population diversity for complex optimization.
• The quantum-inspired PSO is effective for optimizing resource allocation in deep learning-based cloud computing.
• This approach offers better solutions for multi-criteria optimization in cloud environments.