2026· International Journal of Latest Technology in Engineering, Management & Applied Science· Vol 15, pp. 1438-1446· 0 citations
TL;DR
Results show that the proposed framework for thermal-aware and carbon-efficient workload allocation could effectively select out the most suitable servers for workloads not only from viewpoint of carbon usage but also from the thermal perspectives, and at the same time, it also reduces the amount of cooling required and carbon emissions while keeps several performance metrics at comparable levels.
Abstract
As the cloud computing is developing rapidly, the energy consumption of data centers is growing exponentially and has a significant impact on carbon emissions, which imposes great challenges on the sustainable workload management of cloud platforms. At present, most of the workload allocation strategies in cloud platform are made according to some system performance indicators, such as response time, processing ability, server utilization and so on. However, these indicators cannot reflect the carbon consumption of the power supply of the servers. In this paper, we proposed a framework for thermal-aware and carbon-efficient workload allocation which takes several thermal, energy and environmental-related metrics into account for making decisions on allocating workloads to cloud servers. In order to predict the carbon footprint for running a workload on a server, we trained a carbon emission model which is a Gradient Boosting Regressor and it takes a set of attributes into account when predicting the carbon emissions. For any incoming workload, the thermal-aware allocation engine selects the server with minimum predicted carbon emissions by making use of the model learned in training phase and a set of runtime thermal, energy and environmental metrics. Experiments are designed to evaluate the performance and compare with other allocations. Experimental results show that our framework could effectively select out the most suitable servers for workloads not only from viewpoint of carbon usage but also from the thermal perspectives, and at the same time, it also reduces the amount of cooling required and carbon emissions while keeps several performance metrics at comparable levels. Our work shows that by taking into account the thermal metrics and the current carbon intensity information, we could make sustainable scheduling decisions in cloud platforms.
The analysis proves that the combination of carbon awareness and reinforcement learning helps to create an intelligent, adaptive, and ecologically sustainable system for managing cloud resources.
Kirupavathy P., Hareeni C., Jayashri K. et al.· Journal of Ubiquitous Comput...· 0 citations
Data centers in the cloud use a lot of energy and produce considerable carbon emissions because of the growing requirements for computational and AI-heavy processes. Currently, available cloud schedulers pay attention to performance and resource usage optimization without much consideration of carbon footprint efficiency, security compliance, and region-based regulations for cloud data centers. In this research paper, we suggest creating a Secure Carbon-Aware Scheduler that includes machine learning-based carbon intensity prediction, prioritization of the workload, and compliance-based scheduling for distributed cloud environments. Our solution includes using a hybrid approach to carbon monitoring by integrating live carbon intensity data from one pilot region with predicted data based on machine learning models for distributed regions. We trained three ML algorithms such as Random Forest, Gradient Boosting, and Linear Regression for three years of data on carbon intensity, and the best models were chosen according to R2 score and MAE metric.The scheduler performs an assessment of potential regions through a multi-criteria objective scoring function that is based on carbon footprint, security compliance, workload prioritization, and system performance. The architecture employs security enforcement methods that support policies such as GDPR and HIPAA, as well as asymmetric cryptography-based security features for sensitive workloads. Experiments have been conducted by simulating cloud workloads from fourteen different regions and showed promising results in terms of workload prioritization and better carbon-aware scheduling when compared to conventional scheduling algorithms.
Shreenidhi Anant Bhat, P. M.· 2026 4th International Confe...· 0 citations
The rapid expansion of cloud computing and large-scale data centers has significantly increased energy consumption and carbon emissions, creating critical sustainability concerns for modern computing infrastructures. This paper proposes the Adaptive Carbon-Aware Virtualized Energy-efficient Scheduling (ACAVES) framework to improve resource utilization and reduce environmental impact in cloud environments. The framework combines workload monitoring, task classification, virtual machine consolidation, carbon-aware scheduling, and energy optimization within an integrated architecture. An adaptive scheduling mechanism allocates workloads according to utilization patterns, energy requirements, and carbon emission estimates. Experimental evaluation was performed using heterogeneous workloads containing 10,000 tasks executed over 50 physical servers and 200 virtual machines. Results demonstrate that the proposed ACAVES framework reduced energy consumption from 520 kWh to 385 kWh and carbon emissions from 310 kgCO2 to 215 kgCO2. Additionally, server utilization improved from 68% to 87%, while average task completion time decreased from 820 ms to 670 ms, confirming the effectiveness and scalability of the proposed sustainable scheduling framework.
S. K, Kishore Bitra, Usha Desai· 2026 International Conferenc...· 0 citations
There are concerns over cloud computing infrastructure as whether it could be sustainable for exponential growth of
AI workloads. In this era, the present research examines the trends on energy consumption on cloud data centers and proposes
that the use of energy efficient modules can help reduce the operational energy usage by approximately 70% without
compromising the quality of service. This study suggests a holistic approach to green computing in the cloud for AI applications,
which includes smart resource management, optimization of hardware and the use of modern cooling technologies. Based on
industry facts, empirical evidence and key cloud provider reports, this study examines the priority issues in the current
infrastructure and suggests solutions to these issues on the basis of empirical evidence. It is uncovered in the findings that more
than 15% of energy is used by AI data centers and they are projected to consume up to 50% by 2030. It is clear there is a high
level of consumption and fast action is required. This study evaluates the benefits of dynamic allocation of resources,
virtualization, specialized AI accelerators, energy-aware scheduling and adoption of green energy. The study results indicate that
the results of using several techniques in combination along with integrated techniques yield excellent results compared to
interferences separately. The performance of a TPU based system is better than a general-purpose GPU, and AI-based cooling
technology can reduce electricity consumption by approximately 40%. In addition, from on-premise to optimized cloud structure,
migration can minimize carbon emission by 80%. This research contributes valuable, concrete and theoretical strategies for
sustainable application of AI in cloud data centers, fulfilling a growing demand for computing.
S. Gudi· International Journal for Re...· 0 citations
Distributed batch data processing applications are widely executed on cloud-based resources where restricted user access to node-level hardware energy counters hinders transparent sustainability accounting. Energy and carbon attribution methodologies therefore depend on power models and available resource utilisation traces, yet the accuracy of these estimates has to be validated while direct counters are available. In this work, we use Apache Spark running on Kubernetes as a case-study dataflow runtime and cluster resource manager to compare model-based energy estimates to Intel RAPL package and DRAM energy on an AWS bare-metal cloud and an on-premises cluster, comparing different CPU usage signals and memory coefficients. We show that external monitoring improves signed package-energy error relative to Spark task traces, reducing underestimation from -29.58% to -24.41% on AWS and from -24.00% to -16.22% on-premises.
Youssef Moawad, Kathleen West, Vasilis Bountris et al.· 0 citations
: Green Cloud Computing has become a sustainable computing paradigm which seeks to minimize energy usage, carbon dioxide emissions and the cost of operating systems while keeping performance peak in Cloud Computing systems with a high level of services. Clustering of all cloud computing, big data analytics, Artificial Intelligence (AI) is putting enormous pressure on the world's economy and the environment to supply the energy needed for its escalating growth. Efficient use of the computer resources such as job scheduling, load balancing and fault tolerance have become essential to maximize cloud operations and ensure sustainable utilization of computer resources. Task scheduling allows to schedule tasks to the right resources to minimize execution time and energy, load balancing allows to distribute the tasks equally between servers to avoid hot spots and optimize the use of resources. To ensure service availability and system reliability, fault-tolerance mechanisms offer fault detection, prediction and recovery in the event of hardware, software and network failure. To solve these challenges, the Deep Learning (DL) techniques have been found to be effective solutions for the intelligent workload prediction, adaptive resource allocation, proactive fault detection and automatic decision making. Cloud systems can be designed to be environmentally friendly while providing good service quality and performance by integrating carbon emission factors with scheduling, load balancing and fault-tolerant decision making. Combining DL models with the job scheduling, load balancing and fault-tolerance features of the cloud computing systems, cloud service providers can optimize even further energy consumption, lower carbon footprint, enhance Quality of Service (QoS) and improve service dependability. Optimization algorithms, Reinforcement Learning (RL), model-based methods, Neural Networks and hybrid intelligent approaches have been the recent developments with great promise for developing sustainable and resilient cloud computing environments. This article discusses several cloud-based research efforts and offers a thorough overview and comparison of current strategies for task scheduling that is both energy efficient and aware of carbon emissions. By systematically reviewing existing resource management techniques, this study highlights critical gaps and opportunities for improvement. Researchers interested in innovative cloud computing and discovering ways to lower emissions of carbon in cloud settings will find the results useful.
M. Dhanalakshmi, S. Manoharan· International Journal of Sci...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
