Balanced Energy-Aware and Fault-Tolerant Data Center Scheduling

Area of Science:

• Computer Science
• Network Engineering
• Green Computing

Background:

• Data center networks face increasing demands from diverse internet applications.
• Energy consumption and fault tolerance are critical research areas in large-scale networks.
• Existing approaches often present a tradeoff between energy efficiency and fault tolerance.

Purpose of the Study:

• To propose a novel approach for achieving a better tradeoff between energy efficiency and fault tolerance in data center networks.
• To introduce the Energy-Aware Fault-Tolerant (EAFT) approach for resource scheduling.
• To reduce operating costs in data centers while ensuring high availability.

Main Methods:

• Development of the Energy-Aware Fault-Tolerant (EAFT) data center network scheduler.
• Scheduling resources for energy efficiency while maintaining one level of redundancy for fault tolerance.
• Comparative analysis of EAFT with existing energy-efficient resource scheduling techniques.

Main Results:

• The EAFT approach demonstrates a favorable balance between energy efficiency and fault tolerance.
• Analysis of workload distribution, average tasks per server, and energy consumption under EAFT.
• Evaluation of the impact of energy efficiency techniques on overall data center network performance.

Conclusions:

• The proposed EAFT approach offers an effective solution for energy-efficient and fault-tolerant data center networks.
• EAFT achieves reduced operating costs through optimized resource scheduling and redundancy management.
• This study highlights the importance of considering both energy efficiency and fault tolerance for sustainable data center operations.