QoS-Aware Algorithm Based on Task Flow Scheduling in Cloud Computing Environment

Area of Science:

• Computer Science
• Artificial Intelligence
• Operations Research

Background:

• Efficient task scheduling is crucial for cloud computing and edge devices.
• Minimizing energy consumption while adhering to Quality of Service (QoS) is a significant challenge.
• Existing scheduling algorithms often struggle to balance resource utilization and QoS adherence.

Purpose of the Study:

• To develop an energy-efficient task scheduling model.
• To improve the accuracy of Quality of Service (QoS) prediction.
• To optimize task scheduling policies using hybrid metaheuristic algorithms.

Main Methods:

• Utilized an ARIMA prediction model optimized with a Kalman filter for QoS state analysis.
• Developed a hybrid Particle Swarm Optimization (PSO) and Gravitational Search Algorithm (GSA) for scheduling policy calculation.
• Proposed an enhanced hybrid PSO (HPSO) algorithm for task scheduling.

Main Results:

• The HPSO algorithm demonstrated a 16.51% greater reduction in resource consumption compared to the original hybrid algorithm.
• The optimized prediction model led to a 0.053% decrease in Service-Level Agreement (SLA) violations.
• The proposed method effectively balances energy efficiency and QoS requirements.

Conclusions:

• The hybrid PSO-GSA approach, enhanced with optimized QoS prediction, offers superior performance in energy-efficient task scheduling.
• Accurate QoS prediction is vital for effective resource management and SLA compliance.
• This research provides a robust framework for optimizing computational resource allocation in complex systems.