Optimal robust configuration in cloud environment based on heuristic optimization algorithm.

Area of Science:

• Cloud Computing Performance Analysis
• System Robustness Engineering
• Heuristic Optimization Algorithms

Background:

• Cloud computing systems are susceptible to unpredictable perturbations causing performance degradation.
• Existing research often prioritizes profit maximization or waiting time minimization, neglecting performance degradation.
• Defining acceptable performance thresholds is crucial for maintaining system stability.

Purpose of the Study:

• To quantify the impact of perturbations on cloud computing performance.
• To develop a robust configuration strategy for server size and speed.
• To introduce a method for measuring system robustness against perturbations.

Main Methods:

• Defining a feasible region based on minimum acceptable profit and maximum acceptable waiting time.
• Utilizing the concept of robustness to guide server configuration.
• Proposing and evaluating a heuristic optimization algorithm for robustness measurement.

Main Results:

• The proposed heuristic optimization algorithm demonstrates high accuracy.
• The algorithm's solution magnitude error is on the order of 10^-6 compared to benchmark schemes.
• The strategy effectively maintains cloud system performance at an acceptable level under perturbation.

Conclusions:

• The developed robustness strategy effectively mitigates performance degradation in cloud computing.
• The proposed heuristic optimization algorithm offers a precise method for robustness measurement.
• This approach provides a reliable framework for configuring cloud systems to withstand perturbations.