Energy Efficiency Of Data Centers Research Articles

Deep machine learning-based power usage effectiveness prediction for sustainable cloud infrastructures

The expansion of online services, the advent of big data, and the development of Internet of Things (IoT) technology have led to an exponential growth in the number of data centers (DC). Within this new digital systems, DCs are the keystone. This poses major environmental problems through high energy consumption. The energy consumption growth of the DC’s systems are expected to account for 3.2% of global carbon emissions by 2025. This remarkable growth in the energy consumption of DCs represents today a major challenge for energy efficiency. To meet these challenges, Power Usage Effectiveness (PUE) can be used to measure and optimize DC energy efficiency. Thus, accurate PUE prediction is challenging. Moreover, the non-linearity of the different DCs subsystems interactions and their inter-dependencies make the prediction of PUE more difficult since it is affected by complex factors such as workload variation, in addition to weather conditions and humidity, etc.In this paper, we present three machine learning models, MLP, Resilient Backpropagation-based DNN and Attention-based LSTM, to predict PUE values. The performance of these models has been evaluated using two datasets gathered from two case studies of real DCs. For both DC cases, experimental results showed that the attention-based LSTM model outperforms resilient backpropagation-based DNN and MLP models in terms of prediction errors and PUE.

CUCKOO-ANN Based Novel Energy-Efficient Optimization Technique for IoT Sensor Node Modelling

Wireless sensor networks (WSNs) based on the Internet of Things (IoT) are now one of the most prominent wireless sensor communication technologies. WSNs are often developed for particular applications such as monitoring or tracking in either indoor or outdoor environments, where battery power is a critical consideration. To overcome this issue, several routing approaches have been presented in recent years. Nonetheless, the extension of the network lifetime in light of the sensor capabilities remains an open subject. In this research, a CUCKOO-ANN based optimization technique is applied to obtain a more efficient and dependable energy efficient solution in IoT-WSN. The proposed method uses time constraints to minimize the distance between sources and sink with the objective of a low-cost path. Using the property of CUCKOO method for solving nonlinear problem and utilizing the ANN parallel handling capability, this method is formulated. The resented model holds significant promise since it reduces average execution time, has a high potential for enhancing data centre energy efficiency, and can effectively meet customer service level agreements. By considering the mobility of the nodes, the technique outperformed with an efficiency of 98% compared with other methods. The MATLAB software is used to simulate the proposed model.

Cleaner, greener data centres

Abstract John Booth MBCS, Data Centre Energy Efficiency and Sustainability Consultant at Carbon3IT, explores the detrimental trajectory of data centre energy use, against a backdrop of COP26, climate change and proposed EU directives.

Virtual Machine Consolidation using Roulette wheel selection Strategy

Cloud computing data centers are growing rapidly in both number and capacity to meet the increasing demands for highly-responsive computing and massive storage. Such data centers consume enormous amounts of electrical energy resulting in high operating costs and carbon dioxide emissions. The reason for this extremely high energy consumption is not just the quantity of computing resources and the power inefficiency of hardware, but rather lies in the inefficient usage of these resources. Virtual Machine [VM] consolidation involves live migration of VMs hence the capability of transferring a VM between physical servers with a close to zero down time. It is an effective way to improve the utilization of resources and increase energy efficiency in cloud data centers. VM consolidation consists of host overload/under load detection, VM selection and VM placement. In Our Proposed Model We are going to use Roulette-Wheel Selection Strategy, Where the VM selects the Instance type and Physical Machine [PM] using Roulette-Wheel Selection Mechanism Keywords—searchable encryption, dynamic update, cloud computing

Analysis of methods for assessing the energy efficiency of data centers using the power usage effectiveness method

The article deals with the analysis of methods for assessing the energy efficiency of data centers according to the Power Usage Effectiveness method. The demand for data centers which consumes a large amount of electricity is growing with the growth of digitalization and the accumulation of big data in the network. The energy consumption of the cooling system for the machine room accounts for a significant part of the operating costs of the building. Free cooling in a refrigeration system reduces energy consumption much more than operating systems with a vapor-compression cycle. In 2006 according to The Green Grid, the assessment method of Power Usage Effectiveness has become an international standard for measuring energy efficiency and is widely used in the design and operation of data centers. In this regard, the operation principles of free-cooling chillers are considered. The calculation example of the system payback in free-cooling is also given.

Research Status and Future Development of Cooling Technologies for Green and Energy-Efficient Data Centers

Data centers are important infrastructure in China, with constantly increasing industry size, market revenues, and electricity consumption. The major cooling systems of data centers currently in operation in China are air-cooled, with a low energy utilization rate and great potentials for energy saving. Therefore, it is imperative to develop novel cooling technologies for green and energy-efficient data centers. The cooling technology needs to solve two problems. First, as equipment heat production increases, the cooling capacity of the cooling system needs to be improved to match heat production and transfer rates. Second, to address the low energy efficiency problem, green and energy-efficient cooling technologies with industrial application prospects need to be developed. This study reviews the development status of data center cooling methods and categorizes cooling technologies into air conditioner cooling, fresh air cooling, indirect cooling, direct single-phase liquid cooling, and direct two-phase liquid cooling. We analyze the development trends of cooling technologies and suggest that green and high-efficient data centers should focus on the improvement in the efficiency of cooling systems, optimizations of hot and cold fluid channels, applications of new liquid cooling materials, and promotion of heat recovery systems. Furthermore, we propose suggestions for the stable and orderly development of the data center industry from the aspects of optimizing top-level design, developing key technologies, formulating scientific standards, improving industrial layout, and focusing on energy structure.

MAGNETIC: Multi-Agent Machine Learning-Based Approach for Energy Efficient Dynamic Consolidation in Data Centers

Improving the energy efficiency of data centers while guaranteeing Quality of Service (QoS), together with detecting performance variability of servers caused by either hardware or software failures, are two of the major challenges for efficient resource management of large-scale cloud infrastructures. Previous works in the area of dynamic Virtual Machine (VM) consolidation are mostly focused on addressing the energy challenge, but fall short in proposing comprehensive, scalable, and low-overhead approaches that jointly tackle energy efficiency and performance variability. Moreover, they usually assume over-simplistic power models, and fail to accurately consider all the delay and power costs associated with VM migration and host power mode transition. These assumptions are no longer valid in modern servers executing heterogeneous workloads and lead to unrealistic or inefficient results. In this paper, we propose a centralized-distributed low-overhead failure-aware dynamic VM consolidation strategy to minimize energy consumption in large-scale data centers. Our approach selects the most adequate power mode and frequency of each host during runtime using a distributed multi-agent Machine Learning (ML) based strategy, and migrates the VMs accordingly using a centralized heuristic. Our Multi-AGent machine learNing-based approach for Energy efficienT dynamIc Consolidation (MAGNETIC) is implemented in a modified version of the CloudSim simulator, and considers the energy and delay overheads associated with host power mode transition and VM migration, and is evaluated using power traces collected from various workloads running in real servers and resource utilization logs from cloud data center infrastructures. Results show how our strategy reduces data center energy consumption by up to 15 percent compared to other works in the state-of-the-art (SoA), guaranteeing the same QoS and reducing the number of VM migrations and host power mode transitions by up to 86 and 90 percent, respectively. Moreover, it shows better scalability than all other approaches, taking less than 0.7 percent time overhead to execute for a data center with 1,500 VMs. Finally, our solution is capable of detecting host performance variability due to failures, automatically migrating VMs from failing hosts and draining them from workload.

An Eco-Friendly Approach for Reducing Carbon Emissions in Cloud Data Centers

Based on the Saudi Green initiative, which aims to improve the Kingdom's environmental status and reduce the carbon emission of more than 278 million tons by 2030 along with a promising plan to achieve net-zero carbon by 2060, NEOM city has been proposed to be the “Saudi hub” for green energy, since NEOM is estimated to generate up to 120 Gigawatts (GW) of renewable energy by 2030. Nevertheless, the Information and Communication Technology (ICT) sector is considered a key contributor to global energy consumption and carbon emissions. The data centers are estimated to consume about 13% of the overall global electricity demand by 2030. Thus, reducing the total carbon emissions of the ICT sector plays a vital factor in achieving the Saudi plan to minimize global carbon emissions. Therefore, this paper aims to propose an eco-friendly approach using a Mixed-Integer Linear Programming (MILP) model to reduce the carbon emissions associated with ICT infrastructure in Saudi Arabia. This approach considers the Saudi National Fiber Network (SNFN) as the backbone of Saudi Internet infrastructure. First, we compare two different scenarios of data center locations. The first scenario considers a traditional cloud data center located in Jeddah and Riyadh, whereas the second scenario considers NEOM as a potential cloud data center new location to take advantage of its green energy infrastructure. Then, we calculate the energy consumption and carbon emissions of cloud data centers and their associated energy costs. After that, we optimize the energy efficiency of different cloud data centers’ locations (in the SNFN) to reduce the associated carbon emissions and energy costs. Simulation results show that the proposed approach can save up to 94% of the carbon emissions and 62% of the energy cost compared to the current cloud physical topology. These savings are achieved due to the shifting of cloud data centers from cities that have conventional energy sources to a city that has rich in renewable energy sources. Finally, we design a heuristic algorithm to verify the proposed approach, and it gives equivalent results to the MILP model.

Data centre energy efficiency optimisation in high-speed packet I/O frameworks

To meet the demands for faster network packet processing requirements in data centres, packet I/O accelerator solutions have emerged which are dedicated to enabling faster packet forwarding capabilities. Often, such I/O frameworks are implemented in software that is hosted on a Linux operating system and for many of the solutions the software is run in the system's userspace, thus bypassing the operating system's kernel to facilitate a streamlined packet throughput rate. However, operating outside of the kernel loses the real-time interrupt facility, so many of the I/O accelerator schemes rely on a very high-speed polling solution to react to receive packet traffic arrivals; the continuous fast polling rate results in the CPU running hot and wasting energy. An experimental approach to reduce such energy inefficiencies is introduced based on an adaptive polling mechanism (APM). The investigation results show potential energy CPU savings ranging between 9% and 76%.

Probability-Based Online Algorithm for Switch Operation of Energy Efficient Data Center

The huge amount of energy consumed by the data centers around the world every year motivates the cloud service providers to operate data centers in a more energy efficient way. A promising solution is to turn off the idle servers, which, however, may be turned on later, incurring a significant startup cost. The problem turns to dynamically provisioning the workload, and cutting down the energy cost which includes the power to support the running of data center and the startup cost. Different from previous studies that usually consider the worst case performance guarantee when designing online algorithms, this paper considers the average case which is more practical. We propose a simple online algorithm based on the expectation of job interval of workload, which is proven to be optimal for exponential and uniform distributions and achieves tight competitive ratio $\frac{e}{e-1}$ and $\frac{4}{3}$ respectively for them. Simulations using the synthetic data verify our theoretical analysis. Numerical results employing Google's data center workload trace demonstrate that the proposed algorithm outperforms the worst case based algorithm in terms of operation cost reduction.

Energy efficient VM placement and task scheduling techniques to achieve green cloud environment

Continuous increase in the demand for cloud services has resulted in increased workloads. Effective distribution of these augmented tasks on a limited infrastructure for providing the services is a challenging issue. This necessitates servers to run continuously raising the power usage in the data centres. High power consumption results in releasing major amount of CO2 gas to the atmosphere, polluting our natural environment. So, there is a need to think of green cloud computing to reduce this effect of harmful gas to our environment. Designing an energy efficient data centre is a perennial challenge. Hence in this paper, optimal virtual machine placement strategy using knapsack dynamic programming technique and an efficient precedence-based task distribution approach are proposed to alleviate the power consumption and achieve the quality of service maintaining the service level agreements. The implementation results obtained have proved its power efficiency with better quality of service.

Datacentre energy efficiency optimization in high-speed packet I/O frameworks

Datacentre energy efficiency optimization in high-speed packet I/O frameworks

Applying a Model of Technology Diffusion to Quantify the Potential Benefit of Improved Energy Efficiency in Data Centres

Data centres are a key infrastructure for the global digital economy, helping enable the EU “Digital Decade” by 2030. In 2015, data centres were estimated to consume 2.5% of EU electricity demand. In Ireland, the concentrated presence of data centres could consume 37% of national electricity demand by 2028. The uncertainty of data centre facility-level energy efficiency paired with the need to achieve a low-carbon economy pose significant challenge for generation and transmission network planning. This is the first paper to apply a model of technology diffusion with a national forecast of changes in Irish data centre electricity demand through more efficient liquid cooling. The methodology serves as a technology-agnostic resource for practitioners performing forecasts under uncertainty with limited information. Results suggest that technology adoption could lower national electricity demand by 0.81% if adopted by new plant from 2019 to 2028. Savings rise to 3.16% over the same period if adopted by new and existing data centres. Adoption would also lower related emissions by 4.70% and 23.04% over the same period across both scenarios, respectively. Results highlight substantial potential electricity and associated emissions savings available in the sector and suggest policy options to support a transition towards a low-carbon economy.

Enhancing the Security and Performance of Cloud for E-Governance Infrastructure

E-Governance is getting momentous in India. Over the years, e-Governance has played a major part in every sphere of the economy. In this paper, we have proposed E-MODI (E-governance model for open distributed infrastructure) a centralized e-Governance system for government of India, the implementation of this system is technically based on open distributed infrastructure which comprises of various government bodies in one single centralized unit. Our proposed model identifies three different patterns of cloud computing which are DGC, SGC and CGC. In addition, readiness assessment of the services needs to migrate into cloud. In this paper, we propose energy efficient VM allocation algorithm to achieve higher energy efficiency in large scale cloud data centers when system on optimum mode. Our objectives have been explained in details and experiments were designed to demonstrate the robustness of the multi-layered security which is an integration of High secure lightweight block cipher CSL along with Ultra powerful BLAKE3 hashing function in order to maintain information security triad.

HUNTER: AI based holistic resource management for sustainable cloud computing

The worldwide adoption of cloud data centers (CDCs) has given rise to the ubiquitous demand for hosting application services on the cloud. Further, contemporary data-intensive industries have seen a sharp upsurge in the resource requirements of modern applications. This has led to the provisioning of an increased number of cloud servers, giving rise to higher energy consumption and, consequently, sustainability concerns. Traditional heuristics and reinforcement learning based algorithms for energy-efficient cloud resource management address the scalability and adaptability related challenges to a limited extent. Existing work often fails to capture dependencies across thermal characteristics of hosts, resource consumption of tasks and the corresponding scheduling decisions. This leads to poor scalability and an increase in the compute resource requirements, particularly in environments with non-stationary resource demands. To address these limitations, we propose an artificial intelligence (AI) based holistic resource management technique for sustainable cloud computing called HUNTER. The proposed model formulates the goal of optimizing energy efficiency in data centers as a multi-objective scheduling problem, considering three important models: energy, thermal and cooling. HUNTER utilizes a Gated Graph Convolution Network as a surrogate model for approximating the Quality of Service (QoS) for a system state and generating optimal scheduling decisions. Experiments on simulated and physical cloud environments using the CloudSim toolkit and the COSCO framework show that HUNTER outperforms state-of-the-art baselines in terms of energy consumption, SLA violation, scheduling time, cost and temperature by up to 12, 35, 43, 54 and 3 percent respectively.

Slow Replica and Shared Protection: Energy-Efficient and Reliable Task Assignment in Cloud Data Centers

With the explosive growth in the scale of cloud computing infrastructures, reliability and energy efficiency have become important concerns considering the great complexity of cloud data centers. There is an urgent need for efficient task assignment that can dispatch tasks to appropriate cloud data center servers, which is critical to achieve reliability and energy efficiency in current cloud data centers. Most of the research on task assignment focuses on only one of the objectives of reliability and energy efficiency, while the two objectives are intrinsically conflicting with each other. In this paper, we deal with the problem of task assignment in data centers, with the objective of minimizing the energy consumption while providing failure tolerance to task execution failure. We propose a reliability-aware and energy-efficient task replica assignment algorithm based on running task replicas at a low speed and enabling multiple task replicas to share the same server resources. Each task in a job processed by the cloud computing platform has two instances: main task and task replica (shadow). Each main task runs on an individual server, and the task replica associated with the main task is assigned on a different server. The main tasks run at the full server speed, while the task replicas run at a lower rate than the main tasks. The task replicas can be mapped onto dedicated backup servers or be assigned to the servers on which the main tasks are running. Multiple task replicas can share the same server resources to reduce the number of servers required. We conduct experiments through simulations. Experimental results demonstrate that the proposed algorithm can effectively reduce the energy consumption, while achieving a good balance between the number of servers used and job completion time.

A joint energy efficiency optimization scheme based on marginal cost and workload prediction in data centers

With the widespread development of cloud computing, the dramatic increase in the number and size of data centers (DCs) has resulted in substantial energy consumption and serious environmental problems. This creates a challenge for the further development of DCs. It is imperative to improve the overall energy efficiency of DCs. According to the statistics, servers and cooling systems are the main energy-consuming components of DCs. Recently, several energy-efficient strategies have been developed to address these problems. However, most of these works only consider the energy optimization of servers or cooling systems separately. Therefore, a Joint Energy Efficiency Optimization Scheme (JEES) is proposed in this paper, where the energy consumed by servers and the cooling system is jointly considered, and coordinately optimized. JEES includes a dynamic online task scheduling algorithm based on marginal cost evaluation, a resource management strategy that integrates the workload prediction technique to manage resources, and a task migration method using marginal cost evaluation. By using the proposed techniques, the total energy consumption of DCs can be reduced. Extensive experiments have been conducted based on real-world workload traces, and the results demonstrate that compared with other techniques, the proposed scheme effectively improves the overall resource utilization and reduces the total energy consumption of DCs.

Optimal schemes and benefits of recovering waste heat from data center for district heating by CO2 transcritical heat pumps

Recovering waste heat from data center (DC) for district heating by CO2 transcritical heat pumps can effectively improve the performance of DC and reduce CO2 emission of district heating. However, the optimal design schemes, financial and environmental benefits, and market competitiveness of waste heat heating system are still unclear for different application scenarios. To realize the optimal system design and evaluate the benefits at different application scenarios, this work analyzed and compared the comprehensive performance of four system design schemes, considering two waste heat recovery locations and two cycle types. The maximum coefficient of performance (COP) was set as the optimization goal. Influences of electricity and heat prices on the optimal scheme and thermo-economic performance of waste heat heating system were analyzed. The direct electric-heating, coal-heating, gas-heating, air source heat pump, and ground source heat pump were selected as the comparative heating methods to evaluate the market competitiveness of waste heat heating system. Results show that using waste heat of cooling water from IT room achieves a better thermo-economic performance than that from chillers, which increases maximum COP by 18.2%–28.9% and reduces system investment cost by 4.2%–10.2%. The COP of IHE cycle (e.g., add an internal heat exchanger in the simple cycle) is larger, whereas the simple cycle has a lower investment cost. The cycle type with the shortest dynamic payback period depends on actual electricity and heat prices. The financial and environmental benefits of waste heat heating system are very attractive since it can reduce energy cost by 23.0%–75.0% compared with common heating methods and reduce CO2 emission by 12,880 tons annually compared with gas-heating. Furthermore, the energy efficiency of DC will be improved by waste heat reuse, and the annual energy reuse effectiveness (ERE) can decrease from 1.296 to 0.902 at the annual heating time of 121 days.

Sustainability assessment and emergy analysis of employing the CCHP system under two different scenarios in a data center

The high energy consumption of the data centers is a major challenge in sustainable development. In addition to the energy consumption of Information Technology (IT) section in a data center, the energy consumed in the cooling section for eliminating the heat generated by these centers is also significant. For this reason, finding an effective solution to decrease the energy consumption and to increase the energy efficiency in data centers has recently given much attention by researchers. In this paper, the combined cooling, heating and power (CCHP) system is employed in a data center based on absorption chiller scenario and compression chiller scenario. The proposed CCHP systems for the data center are evaluated in case of sustainability using emergy analysis taking into account non-renewable/renewable environmental resources and non-renewable/renewable purchased inputs as well as the other environmental viewpoints. Results show that the total emergy consumption using the first and second scenario is 4.02 E+18 and 6.01 E+18 solar emjoule, respectively. It was found that the maximum emergy consumption in the first scenario is related to absorption chiller by 3.49 E+18 solar emjoule and the maximum emergy consumption in second scenario is related to compression chiller by 5.24 E+18 solar emjoule. In addition, using the first and second scenario, the sustainability index is obtained 1.70 and 2.50 respectively, which indicates that the second scenario is more sustainable than the first one.

Increasing the energy efficiency of a data center based on machine learning

AbstractEnergy efficiency of data centers (DCs) is of great concern due to their large amount of energy consumption and the foreseeable growth in the demand of digital services in the future. The past decade witnessed improvements of the energy efficiency of DCs from an extensive margin—a shift from small to large, more efficient DCs. Improvements from the intensive margin, that is, from more efficient operation, would be critical in limiting the energy consumption and environmental impact of DCs in the upcoming period. Machine learning algorithms have advantages in optimizing the operation of DCs to improve energy efficiency as they have shown the potential of discovering control strategies not found by traditional method, and producing working condition‐dependent control strategies. This study proposes ready‐to‐use machine learning methods with practical details to decrease the most commonly used energy efficiency metric—power usage effectiveness in DCs. We achieved an accurate prediction by properly selecting the features used in the proposed prediction models established by neural network, light gradient boosting machine, recurrent neural network, and random forests. The proposed approaches are implemented in one of the largest hyperscale DCs in China—Tencent Tianjin DC, to optimize the set points of controllable variables in the cooling system and to detect and adjust the unreasonable working conditions in the modular data centers. The lower bound of PUE reduction was 0.005 with the proposed approaches, leading to about 1500 MWh (0.24% of the total designed electricity consumption of this DC) of energy saved per year in this hyperscale DC. The proposed methods have the potential to be transferred to DCs of similar scale, and the framework of our work could serve as a guide for machine learning‐based optimization of environmental indicators in other complex product/service systems.