Green Software Architecture: Carbon-Aware and Energy-Efficient Approaches for Sustainable Cloud Computing

The increasing global reliance on cloud computing has led to a significant rise in energy consumption and carbon emissions from data centers worldwide. In response to this growing environmental concern, the field of Green Software Architecture has emerged as a critical area of research, aiming to minimize the ecological footprint of cloud operations while maintaining optimal system performance. A recent comparative literature review, published in the Spectrum of Engineering Sciences on June 15, 2026, provides a comprehensive overview of carbon-aware and energy-efficient approaches for achieving sustainable cloud computing. The review meticulously examined twenty distinct research studies, categorizing them into various sustainability domains. These domains include carbon-aware scheduling, which focuses on optimizing workload distribution to leverage periods of lower carbon intensity electricity; energy-efficient resource management, involving techniques to reduce power consumption of hardware and software; renewable energy integration, exploring how cloud infrastructure can be powered by sustainable sources; AI-based optimization, utilizing artificial intelligence to enhance efficiency; and green software design, promoting the development of inherently eco-friendly applications. The findings from this extensive review underscore several key strategies for advancing cloud sustainability. Notably, the implementation of intelligent workload scheduling, which dynamically adjusts computing tasks based on energy availability and carbon intensity, proves highly effective. Furthermore, the integration and increased utilization of renewable energy sources for powering data centers are identified as paramount. The paper also emphasizes the transformative potential of machine learning optimization, where AI algorithms are employed to predict and manage energy usage more efficiently, leading to substantial reductions in both carbon emissions and overall energy consumption. Beyond highlighting successful strategies, the literature review also addresses the persistent challenges in developing truly environmentally friendly cloud systems. These challenges range from the complexities of integrating diverse renewable energy sources to the need for more sophisticated metrics to accurately measure and report environmental impact. The authors conclude by outlining crucial future research directions, advocating for continued innovation in areas such as advanced predictive analytics for energy management, novel hardware designs for reduced power consumption, and the development of industry-wide standards for green cloud practices. This review serves as a valuable resource for researchers and practitioners committed to fostering a more sustainable future for cloud computing.