Search

In cloud computing environment tasks are allocated among virtual machines (VMs) having different length, starting time and execution time. Therefore, balancing these loads among VM is a key factor. Load balancing has to be carried out in such a manner that all VMs should have balanced to achieve optimal utilization of its capabilities and improve the system performance. In this paper, we proposed a load balancing technique by using modified PSO task scheduling (LBMPSO) to schedule tasks over the available cloud resources that minimizes the makespan and maximizes resource utilization. This is achieved by having proper information among the tasks and resources within the datacenter. Our proposed scheduling algorithm is implemented by using CloudSim simulator. Simulation results clearly shows that proposed scheduling algorithm performs better in reducing makespan and increases the resource utilization than other existing techniques. [ABSTRACT FROM AUTHOR]

In the present paper, we focus on the study of the strongly Menger-edge-connectivity of a server-centric DCN DCell. Let D m , n be the m -dimensional DCell with n -port switches for integers m ≥ 0 and n ≥ 2. Given integers k , m and n. We prove D m , n is ((k − 1) m + (k − 1) n − k 2 + 1) -strongly Menger-edge-connected of order k − 1 for 2 ≤ k < m and n > m ≥ 2 or 2 ≤ k ≤ n and m ≥ n ≥ 2 and D m , n is (m n − m − n + 1) -strongly Menger-edge-connected of order m − 1 for n > m ≥ 2. Also, we show that each of the above results is optimal by using examples. [ABSTRACT FROM AUTHOR]

Green Data Centers (GDCs) are more and more deployed world-wide. They integrate many renewable sources to provide clean power and decrease their operating cost. GDCs are typically deployed over multiple locations where renewable energy availability, bandwidth prices and grid electricity cost have high geographical diversity. This paper focuses on delay-bounded applications in distributed GDCs (DGDCs) and performs cost and energy-effective scheduling of multiple heterogeneous applications verifying delay bound constraints of different tasks. DGDCs' operational cost minimization problem is formulated and successfully optimized using an innovative modified Firefly Algorithm (mFA). Real-life data trace-driven experiments are conducted to evaluate the effectiveness of the proposed mFA in solving this problem. High performance task scheduling results are obtained. The operational cost of each GDC is minimized, the utilization of solar and wind renewable energy from the different geographical locations is maximized while delay bound constraints of all tasks are strictly met. Compared to Bat Algorithm, Simulated-annealing Bat Algorithm and basic firefly algorithm, mFA can produce a schedule that outperforms its peers' drastically in terms of operational cost of DGDCs. Moreover, mFA finds more rapidly both global or local optima than its peers. It succeeds to meet all equality and inequality constraints at all time slots while its peers may sometimes fail to find satisfactory solutions at some particular time slots. [ABSTRACT FROM AUTHOR]

Data centres increase their size and complexity due to the increasing amount of heterogeneous workloads and patterns to be served. Such a mix of various purpose workloads makes the optimisation of resource management systems according to temporal or application-level patterns difficult. Data-centre operators have developed multiple resource-management models to improve scheduling performance in controlled scenarios. However, the constant evolution of the workloads makes the utilisation of only one resource-management model sub-optimal in some scenarios. In this work, we propose: (a) a machine learning regression model based on gradient boosting to predict the time a resource manager needs to schedule incoming jobs for a given period; and (b) a resource management model, Boost, that takes advantage of this regression model to predict the scheduling time of a catalogue of resource managers so that the most performant can be used for a time span. The benefits of the proposed resource-management model are analysed by comparing its scheduling performance KPIs to those provided by the two most popular resource-management models: two-level, used by Apache Mesos, and shared-state, employed by Google Borg. Such gains are empirically evaluated by simulating a hyper-scale data centre that executes a realistic synthetically generated workload that follows real-world trace patterns. [ABSTRACT FROM AUTHOR]

Since the onset of the 21st century, the number of data centers has been increasing continuously, and the energy consumption of these data centers has aroused people's attention. Many experiments have been conducted to reduce the energy consumption in data center cooling, and free cooling utilization based on the mechanically driven two-phase cooling loop (MTCL) system has been proven to be an effective solution for energy saving. This paper reviews the research on and developments in the practical application of the MTCL system for data center cooling in recent decades. It presents a detailed summary of the progress of research on the MTCL system in data centers. The aim is to provide an improved understanding of the system and promote the application of this system in data center energy saving. [ABSTRACT FROM AUTHOR]

Load balancing of tasks on the cloud environment is an important aspect of distributing resources from a data centre. Due to the dynamic computing through the internet; cloud computing agonizes from overloading of requests. Load balancing has to be carried out in such a manner that all virtual machines (VM) should have balanced load to achieve optimal utilization of its capabilities. This paper proposes a novel methodology of dynamic balancing of load among the virtual machines using hybridization of modified Particle swarm optimization (MPSO) and improved Q-learning algorithm named as QMPSO. The hybridization process is carried out to adjust the velocity of the MPSO through the gbest and pbest based on the best action generated through the improved Q-learning. The aim of hybridization is to enhance the performance of the machine by balancing the load among the VMs, maximize the throughput of VMs and maintain the balance between priorities of tasks by optimizing the waiting time of tasks. The robustness of the algorithm has been validated by comparing the results of the QMPSO obtained from the simulation process with the existing load balancing and scheduling algorithm. The comparison of the simulation and real platform result shows our proposed algorithm is outperforming its competitor. [ABSTRACT FROM AUTHOR]

Farmland abandonment has become a worldwide phenomenon, and remote sensing technology has been used to monitor this phenomenon. However, information on only the final net change in abandoned farmland can be obtained based on existing methods, and the various components and contributions of government regulation cannot be identified. These methods therefore cannot meet the needs for the management of abandoned farmland or related policy-adjustments. Therefore, this study uses Longyan Prefecture in China as the study area, and GF-1 satellite images as the main data source, to propose a new rapid monitoring method for abandoned farmland that can reflect and separate the government's regulatory contribution to changes in farmland abandonment. The results show that the abandoned farmland in the study area generally expanded from 2015 to 2019. Despite this observation, government regulation has made some achievements. In 2015–2017 and 2017–2019, 41.72% and 41.67% of the new increased abandoned farmland was offset, respectively, delaying the process of farmland-abandonment. However, the achievement of government regulation decreased by 23.86% in 2017–2019 compared with that during 2015–2017. The results show that our method can reflect the process of farmland abandonment and the contribution of government regulation, can be applied on a large scale, and provides comparable results among regions. Offsetting the newly increased abandoned farmland by improving the contribution of government regulation is an important and direct way to curb farmland abandonment, but special attention should be given to the scientific nature and full implementation of policies in actual management practices. The Chinese GF-1 Satellite images were downloaded from the China Center for Resources Satellite Date and Application website (http://www.cresda.com/EN/). The Advanced Land Observing Satellite (ALOS) data downloaded from the National Aeronautics and Space Administration (NASA) (https://search.asf.alaska.edu/#/). The vector boundary of the research area was obtained from the Chinese Resource and Environment Science and Data Center (https://www.resdc.cn/). Most of the field samples for evaluating the accuracy of the land classification results downloaded from Global Change Data & Discovery (http://www.geodoi.ac.cn/WebEn/doi.aspx?Id=1502). • Various components of abandoned farmland were monitored by remote sensing technology. • Contributions of government regulation to abandoned farmland regulation was identified. • Government's gains have been offset by newly increased abandoned farmland. • Our method can reflect the process of farmland abandonment. [ABSTRACT FROM AUTHOR]

Data centers store the necessary data for the operation of services and applications used by users. They usually consist of many physical servers connected to each other, with virtual machines and other systems, applications, databases installed. In order to increase the reliability of data access and to protect the system from damage or destruction, data from servers are replicated to other centers, servers and instances. The article examines the average replication time for whole virtual machines of various sizes and differential replication of virtual machines for exemplary data center. The research was conducted in order to investigate the influence of bandwidth constraints between data centers on replication times. The article presents sample results from a real system. [ABSTRACT FROM AUTHOR]

Aiming at the power fluctuation and mismatch of the combined cooling, heating, and power (CCHP) system based on proton exchange membrane fuel cells (PEMFCs) and adsorption chiller, this study proposes a multi-stack coupled power supply strategy. The PEMFC stacks are divided into types Ⅰ, Ⅱ, and Ⅲ to meet the electric load and cooling load of the data center, and the heat requirements of the system. Meanwhile, economic analysis is conducted on the single-stack energy supply strategy and the multi-stack coupled energy supply strategy. The results show that with the multi-stack coupling power supply strategy, the cooling power and electric power almost completely match the load of the data center, without power fluctuations and overshoot. By smoothing the PID control results of the current of the stacks-Ⅲ, the heating power fluctuation is significantly reduced, and the maximum overshoot does not exceed 0.5 kW. Therefore, the strategy is conducive to the stable operation of the PEMFC stack and improves the lifetime of the system. Considering investment costs, maintenance costs, hydrogen costs, and electricity benefits, the multi-stack coupled energy supply strategy can save about 6.1 × 105 $ per year. In summary, the multi-stack coupled energy supply strategy has advantages in system lifetime, operational stability, and economy. • Multi-stack coupled power supply strategy is proposed. • The PEMFC are divided into types Ⅰ, Ⅱ, and Ⅲ to meet different power requirements. • The cooling and electric power completely matches the load with the strategy. • The heating power fluctuation is reduced by smoothing the current of stacks-Ⅲ. • The multi-stack coupled energy supply strategy can save about 6.1 × 105 $ per year. [ABSTRACT FROM AUTHOR]

As more people send information to the cloud-fog infrastructure, this brings many problems to the management of computer energy consumption. Therefore, energy consumption management of servers, fog devices and cloud computing platform should be investigated to comply with the Green IT requirement. In this paper, we propose an energy consumption prediction model consisting of several components such as hardware design, data pre-processing, characteristics extraction and selection. Our main goal is to develop a non-invasive meter based on a network of sensors that includes a microcontroller, the MQTT communication protocol and the energy measurement module. This meter measures voltage, current, power, frequency, energy and power factor while a dashboard is used to present the energy measurements in real-time. In particular, we perform measurements using a workstation that has similar characteristics to the servers of a Datacenter locate at the Information Technology Center in ESPOL, which currently provide this type of services in Ecuador. For convenience, we evaluated different linear regression models to select the best one and to predict future energy consumption based on the several measurements from the workstation during several hours which enables the consumer to optimize and to reduce the maintenance costs of the IT equipment. The supervised machine learning algorithms presented in this work allow us to predict the energy consumption by hours and by days. [ABSTRACT FROM AUTHOR]

• A new data center cooling system integrating heat recovery with free cooling is proposed. • The system can work selectively in four different modes according to conditions and demands. • Stable performance of each mode and dynamic performance in mode switching are obtained. • Annual power consumption and heat recovery analysis are conducted in five cities. Rapid growth of data centers has caused a huge increase in energy consumption. Utilizing natural cold resources and harvesting waste heat are two promising ways to reduce the total power consumption of data centers. This paper proposes an integrated system of free cooling and heat recovery. The system has four working modes: mechanical cooling, mechanical heat recovery, pump free cooling and pump heat recovery. Experimental results show that pump free cooling mode can achieve an energy efficiency ratio of 7.99 when outdoor temperature is 2 °C, and different modes can be switched smoothly. The applicability of this system has been evaluated in five different cities in China. The results show that annual average energy efficiency ratio can reach 6.18 and free cooling time can take up as high as 72.9% during operation in Xining. The system can harvest the most heat in Harbin, 7930.96 kW∙ h per year, while the least in Guangdong, 645.31 kW∙ h per year. This means that this new system has relatively great application prospects in northern cold cities of China, while is moderately limited in southern cities. [ABSTRACT FROM AUTHOR]

Spatiotemporal variations of ozone (O 3) taken from the Copernicus Atmosphere Monitoring Service (CAMS) and the second Modern-Era Retrospective Analysis for Research and Applications (MERRA-2) were intercompared and evaluated with ground and ozone-sonde observations over China in 2018 and 2019. Intercomparison of the surface ozone from CAMS and MERRA-2 reanalysis showed significant negative bias (CAMS minus MERRA-2, same below) at Tibetan Plateau of up to 80 µg/m3, and the average R 2 was about 0.6 across China. Evaluated with the ground observations from China National Environmental Monitoring Center (CNEMC), we found that CAMS and MERRA-2 reanalysis were capable of capturing the key patterns of monthly and diurnal variations of surface ozone over China except for the western region, and MERRA-2 overestimated the observations compared to CAMS. Vertically, the CAMS profiles overestimated the ozone-sonde from the World Ozone and Ultraviolet Radiation Data Center (WOUDC) above 200 hPa with the magnitude reaching up to 150 µg/m3, while little bias was found between the reanalysis and observations below 200 hPa. Intercomparison drawn from the vertical distribution between CAMS and MERRA-2 reanalysis showed that the negative bias appeared throughout the troposphere over China, while the positive bias emerged in the upper troposphere and lower stratosphere (UTLS) with high order of magnitude exceeding 100 µg/m3, indicating large uncertainties at higher altitudes. In summary, we concluded that CAMS reanalysis showed better agreement with the observations in contrast to MERRA-2, and the large discrepancy especially at higher altitudes between these two reanalysis datasets could not be ignored. [Display omitted]. [ABSTRACT FROM AUTHOR]

We propose a parallel multi delay (PMD) carrier-assisted differential detection (CADD) receiver for the first time, including PMD asymmetric-CADD (PMD A-CADD) and PMD symmetric-CADD (PMD S-CADD). The transfer function of the PMD CADD receiver is optimized by expanding the number of optical delays to a general selection, thereby suppressing the signal-signal beat interference (SSBI). As a result, the required carrier-to-signal power ratio (CSPR) is reduced, and the optical signal-to-noise ratio (OSNR) sensitivity is drastically improved. Two schemes for further mitigation of residual SSBI are introduced, namely SSBI iterative cancellation algorithm and the increased guard band mitigation method. For SSBI iterative cancellation algorithm, by optimizing the number of optical delays, the optimal CSPR for the P6D A-CADD and the P5D S-CADD can be reduced to −1 dB, the required number of iterations is only 3, and compared with the parallel dual delay-based asymmetric-CADD (PDD A-CADD) receiver, the OSNR sensitivity is improved by about 4.5 dB. For the increased guard band mitigation method, 2.34 GHz and 3.16 GHz guard band are required for the P6D A-CADD and the P5D S-CADD to meet the 7% FEC threshold at 0 dB CSPR. • A parallel multi delay CADD receiver is proposed for data center interconnect applications. • The proposed receiver suppresses SSBI by modifying the transfer function, thereby reducing the optimal CSPR to less than 0 dB. • Two SSBI iteration cancellation algorithms are used to improve the system performance, and the SSBI simplified iteration cancellation algorithm had lower complexity and better performance. [ABSTRACT FROM AUTHOR]

The control of indoor temperature in greenhouses is crucial as it directly impacts the crop's thermal comfort and the performance of heating, ventilation, and air-conditioning (HVAC) systems. Conventional feedback controllers, like on/off, can sometimes make HVAC system work at full capacity when only half that capacity is needed. In contrast, the proportional-integral-derivative (PID) controller, provides precise control based on its P, I, and D parameters. However, it lacks a formal design procedure for optimizing a specified objective function. Previous studies have utilized conventional PID tuning approaches to track room setpoint temperature for residential buildings, data centers, and office buildings, with limited research in greenhouse applications. To address this gap, this study proposes a flexible PID controller that employs a deep reinforcement learning (DRL) algorithm to optimize its parameters, by tracking the setpoints and energy consumption of a greenhouse planted with tomatoes. This approach is different from the typical method of using the trained RL agent directly in HVAC controls. Through a self-made TRNSYS-Python cosimulation framework, the DRL agent interacts directly and in real time with the greenhouse and its plants. Consequently, optimized PID parameters were established and tested in the simulated environment. The resulting performance, in terms of both energy consumption and its ability to maintain the crop's comfort temperature, was compared with the simulated on/off and manually tuned PID controllers. Compared to the on/off baseline control, the proposed PID optimized parameters reduce energy use by 8.81% to 12.99% and the manually tuned PID parameters with the Ziegler-Nichols tuning method reduce energy use by 7.17 %. Additionally, the proposed method had a deviation of 2.07% to 3.13%, while the manually tuned PID controller and the on/off controller had deviations of 7.27% and 3.27%, respectively, from the minimum comfortable temperature. This study serves as a framework for improving the energy efficiency of greenhouse HVAC system operations. [ABSTRACT FROM AUTHOR]

Fifth-generation district heating and cooling (5GDHC) is an emerging concept that exploits heat sharing among prosumers through a low-temperature bidirectional loop. Coordinating the energy interactions in such a system is a complex problem, typically solved through centralized optimization. In addition to privacy issues for prosumers, centralized optimization can prove too hard to solve for large 5GDHC systems and cannot establish a profit distribution between prosumers. To overcome these drawbacks, this paper develops a community-based transactive energy market framework based on distributed optimization of 5GDHC. The optimization is based on the Jacobi-proximal alternating direction method of multipliers. The approach relies on iterative interactions between the network coordinator and the prosumers, the former adjusting the internal price of thermal energy and the latter adapting their heating and cooling demand. A series of test-case 5GDHC networks involving houses, commercial buildings, data centers and central boilers and chillers were numerically simulated to demonstrate how the proposed framework performs, under three electricity tariff structures (constant, time-of-use or pay for peak). Results show that, for the cases tested, the gap between the proposed approach and global optimization is below 5% and that electricity savings compared to a situation without 5GDHC could reach 40%. The proposed method allowed to dynamically adjust the internal price of thermal energy and the prosumers' demand while considering their thermal dissatisfaction. By facilitating 5GDHC operation, this work could help their practical implementation in the future. • 5th generation district heating and cooling (5GDHC) uses heat sharing among prosumers. • New distributed optimization of 5GDHC has below 5% difference with central optimizer. • The new market determines internal price of thermal energy and prosumers' demand. • 5GDHC generates savings of up to 40% under different electricity tariffs. [ABSTRACT FROM AUTHOR]

The power consumption for cooling of data centers increases continuously contributed to the enlargement of heat dissipation demands. Waste heat recovery is regarded as a promising solution for energy conservation and carbon reduction. A compression-assisted absorption refrigeration-heating system aimed at waste heat utilization of liquid-cooling data center is proposed. The effects of heat sources, auxiliary compression process and split ratios on the proposed system performance are discussed, and the analysis of application that synchronously produces chilled water for liquid-cooling data center and hot water for domestic use is conducted. The compression process can provide effective compensation for the system performance that weakens the demands on the quality of cooling source and driving heat source. As the compression ratio raises from 1.067 to 2.0, the minimum generation temperature decreases from 52.0 °C to 27.2 °C. The system can completely satisfy the refrigeration need, of 4.29 kW for liquid-cooling data centers and provide additional thermal energy, of 1.37 kW for heating. A quantitative relationship between the flow split ratios and circulation ratios is found out that can be a wise presentation for the adjustment of the refrigeration and heating loads. This research results can provide theoretical basis and guidance for low-grade waste heat recovery in liquid-cooling data center based on the absorption heat pump technology. • A compression-assisted absorption refrigeration-heating system for waste heat recovery was constructed. • The compression process weakened the demands on the quality of driving heat source and cooling source. • A quantitative relationship between flow split ratios and circulation ratios was found out. • The system satisfied the refrigeration loads of data centers and provided heating effect. [ABSTRACT FROM AUTHOR]

Under increasing heat loads and stricter energy consumption limits for data center servers, traditional heat sinks are becoming inadequate. In this paper, a novel aluminum three-dimensional vapor chamber heat sink of air cooling is developed, in which the evaporator and condenser are connected internally to form a three-dimensional flow channel for working fluid so that two heat transfer methods, phase change and single-phase, are combined in one device. Furthermore, a novel louvered-fin stacked evaporator wick structure and its fabrication process are proposed, which greatly improve thermal performance and temperature uniformity. Thermal characteristics under different heat loads, fan input power, and tilt angles are studied experimentally. Results show that the heat sink has a maximum heat dissipation capability of more than 450W, at which the maximum evaporator surface temperature is within 75 °C. The fan input power can distinctly improve thermal performance, heat load impacts overall thermal resistance slightly and the minimum is 0.083 °C/W. The tilt angle impacts thermal resistance slightly but significantly affects both the evaporator surface and condenser fin temperature uniformity. Under a tilt angle of 90°, regardless of heat loads, the condenser fin temperature is much lower than other tilt angles. • A novel aluminum 3-dimensional vapor chamber heat sink (3DVCHS) is proposed. • A louvered-fin stacked evaporator wick is implemented. • The air-cooled 3DVCHS has amaximum heat dissipation capability of more than 450W. • Heat loads impact the 3DVCHS thermal resistance slightly, the minimum is 0.083 °C/W. [ABSTRACT FROM AUTHOR]

Cold-rolled strips are a pivotal product of advanced strip rolling technology in iron and steel production with stringent process standards. Over an extended period, the existence of 'information silos' between cold strip rolling (CSR) and its upstream processes, particularly hot strip rolling (HSR), causes a persistent challenge in achieving collaborative optimization throughout the entire process, thereby constraining progress in product quality. To overcome this challenge, a digital twin framework was proposed to consolidate diverse data from HSR and CSR production lines into a centralized data centre, forming a 'data continent.' This approach establishes a comprehensive dataset for model-driven training and prediction in virtual space. An improved Light gradient boosting machine (IlightGBM) is introduced, enabling high-precision rolling force prediction in unsteady-state rolling and providing essential variables for establishing the tension optimization objective function. A control function is implemented to minimize rolling force fluctuations, and three population algorithms (TPA) are devised for its solution. This strategy successfully realizes collaborative tension optimization across multiple stands, with the rolling speed acting as a follower. The proposed approach was implemented in a large-scale steel plant equipped with a complete HSR and CSR production line, effectively mitigating flatness defects and thickness deviation in the strip's head section. Thus, the proposed framework is reliable and can be flexibly applied to strip manufacturing processes. [ABSTRACT FROM AUTHOR]

Nowadays, numerous mobile devices (MDs) provide nearly anytime and anywhere services, running on top of various computation-intensive applications. However, bearing limited battery, bandwidth, computing, and storage resources, MDs cannot completely execute all tasks of such applications in real-time. Cloud data centers (CDCs) possess enormous resources and energy, which can help execute tasks offloaded from MDs. Nonetheless, CDCs reside in remote sites, thereby leading to long transmission latency. In recent years, small base stations (SBSs) have emerged to offer close proximity, high bandwidth, and low latency services to their nearby and limited MDs. However, it becomes a new challenge to minimize the total system cost in a complex and heterogeneous architecture. To address it, this work proposes an energy-minimized partial computation offloading technique. First, a limited optimization problem of cost minimization for the system is formulated. Afterward, an improved hybrid meta-heuristic algorithm is developed, which synergistically combines a Metropolis acceptance criterion of simulated annealing and genetic operations. One uniqueness of the proposed algorithm is that it simultaneously determines task allocation among MDs, an SBS, and a CDC, the transmission power of MDs, and bandwidth allocation of wireless channels between MDs and SBS. Experiments with real-life tasks from Google data centers have shown that the proposed G enetic S imulated-annealing-based P article swarm optimization (GSP) significantly achieves lower system cost and faster convergence speed than benchmark peers. • A hybrid system architecture is designed and characterized by triple queueing models. • A constrained total cost minimization problem with multiple constraints is formulated. • The cost minimization problem is solved by a hybrid optimization algorithm named GSP. • It combines simulated annealing and genetic operations into a particle swarm optimizer. • Experiments with realistic traces show GSP achieves lower cost and faster convergence. [ABSTRACT FROM AUTHOR]

Computational Storage Devices, also known as In-Storage computing or In-Suit Processing, offer higher computing power than traditional storage devices. Innovation in computational storage devices is more important than ever because of the exponential growth of digital data produced by both individuals and organizations. With the advent of non-volatile storage devices in mainstream computing (e.g., hard disk drives and solid-state drives), the idea has gained wide appeal, with several academic and commercial prototypes becoming available. This survey aims to provide a systematic overview of the work being done in the area of computational storage and to indicate future directions. This overview considers and examines a number of research questions to comprehensively summarize, analyse and discuss various storage devices such as Hard Disk Drives (HDD), Solid-State Drives (SSD) and Computational Storage Devices (CSDs) in terms of design, programming model, acceleration and energy efficiency. Also, the Load Balancing algorithm has been reviewed from different domains such as cloud computing, data centers and the Internet of Things (IoT). This review not only focuses on the existing literature problems but also requires some effort to explore the future in this area. [ABSTRACT FROM AUTHOR]

The growing demand for cloud computing and storage necessitates the expansion of Data Centers, thereby increasing energy consumption and environmental footprint. The data-driven dynamic thermal control presents a promising solution to mitigate this issue, as it can offer improved thermal control strategies and address the intricate and stochastic nature of the thermal environment. Nonetheless, existing solutions typically exhibit deficiencies in terms of reliability and practicability. To this end, this paper proposes a feasible uncertainty-aware online learning method with reliable dynamics models and controllers. Specifically, Deep Learning models with the capacity for uncertainty quantification are employed to account for the imperfections in pre-trained models, hence providing reliable predictions. The Cross-Entropy Method and Monte Carlo trajectory sampling solve optimization problems while considering uncertainties. It is mathematically analyzed that the uncertainty quantified by the system dynamics model can be captured with sufficient trajectory samples, rendering a reliable controller. A practical framework is proposed and tested on a widely used Data Center model to mitigate the insufficient capacity of Deep Learning models to quantify uncertainties. The results demonstrate its superior performance during the initial training stage, achieving a 66% reduction in violations and 3.76% power savings compared to the default controller. Continuous active online learning enhances its adaptability, resulting in an 80% reduction in violations and a 6% energy saving. The thermal control strategies are thus improved by the proposed new solutions and insights. [ABSTRACT FROM AUTHOR]

• Batch loads are classified into two types based on their adjustable characteristics. • Various flexible resources in the data center are fully utilized. • A new method for wind power scenario generation is proposed. • A sensitivity analysis is conducted on the first type batch processing load. The rapid development of Internet and cloud computing technologies has led to the expansion of the scale of data centers, resulting in a continuous increase in the demand for data processing and storage. This not only results in higher energy consumption but also makes thermal management a key factor of energy management in data centers. A capacity configuration method is proposed for the core thermal management equipment of data center microgrids, electric boilers, cooling systems, and heat pumps. This method not only considers the uncertainty factors of the source and load but also effectively utilizes the characteristics of various flexible resources in data centers, such as the adjustable ability of different types of batch processing loads, air thermal inertia, and waste heat recovery, significantly improving the energy utilization efficiency and enhancing the system's economy and flexibility. In addition, a wind power scenario generation method based on conditional least squares generative adversarial networks was proposed to improve the generation quality of wind power scenarios, and a random optimization model was constructed. Considering a certain province's data center microgrid as a case study, the effectiveness of the model was verified, and a sensitivity analysis of the relevant parameters provided important references for microgrid planning. This study provides a new perspective and tool for the thermal management of data center microgrids and comprehensive utilization of renewable energy, which is of great significance for promoting the improvement of data center energy efficiency and sustainable development. [ABSTRACT FROM AUTHOR]

The virtual power plant (VPP) can aggregate massive distributed resources to participate in power system operation. Different types of resources have been investigated to construct VPPs. Internet data centers (IDCs) are regarded as flexible power consumers that can potentially cooperate with power systems. This paper exhibits the insight for constructing VPPs based on IDCs while comprehensively considering the operational flexibility of the workloads and power supply devices. On the workloads, this paper models the temporal and spatial flexibility of workloads in detail. On the power supply devices, this paper explores their spare capacity with respect to the power supply reliability requirement. Then, a day-ahead scheduling method of the IDC-based VPPs is established based on the operational flexibility modeling. Case studies validate the proposed approach. Moreover, because of the coupling of workload dispatch and power supply device operation, the synergy effect does not equal the sum of its parts. • Exploring the operational flexibility of IDCs to construct VPPs.. • Quantifying the potential of backup power under the reliability requirement. • Realizing the coordination of the backup power and computation workloads in IDCS. • Proposing a day-ahead scheduling framework for IDC-based VPPs. [ABSTRACT FROM AUTHOR]

Cooling of data centers requires a significant amount of energy, comparable to the energy consumption of the servers themselves. The current design of the centralized cooling systems for data centers is based on ideal IT loading conditions (i.e., 100 % loading). However, such conventional design often results in significant oversized cooling systems and leads to substantial energy waste, since most data centers operate at part load in their lifespan. To address this issue, this study proposes an optimal design for centralized cooling systems with multiple chillers under progressive loading. The optimization problem, aimed at minimizing life-cycle cost, is formulated adopting SLSQP (Sequential Least Squares Programming) algorithm. A cooling system model is developed using the manufacturer's performance data of cooling equipment. The optimal designs in different climate zones are identified according to energy performance under full-range loads and ambient temperatures. Furthermore, this study comprehensively analyzes and compares free cooling hours, cooling energy, and life-cycle cost of the optimized designs with conventional designs. The results show that the optimized cooling systems could operate more energy-efficiently, despite decreased free cooling hours (13–860). Significant cooling energy savings over the lifespan could be achieved, i.e., 4–22 %, corresponding to the PUE reductions of 0.02–0.11, depending on climate conditions and control strategies. [ABSTRACT FROM AUTHOR]

In the realm of cloud computing, effective resource allocation can significantly enhance the energy efficiency of datacenters. Task scheduling and Virtual Machine Placement (VMP) are two pivotal aspects of resource allocation. However, in current research, they are often treated separately, overlooking the potential for integrated optimization. In this paper, we propose an integrated solution for task scheduling and VMP in energy-efficient datacenters, based on queueing theory and Deep Reinforcement Learning (DRL) methods. This novel and comprehensive approach provides an alternative perspective for resource scheduling strategies in datacenters. We construct a queueing theory model for task scheduling, aiming to minimize the number of VMs that need to be instantiated, while ensuring that Service Level Agreement (SLA) violation remains at a low level. Furthermore, we design a VMP algorithm based on DRL for real-time selection of Physical Hosts (PHs) for deploying VMs. Finally, we conduct a simulation evaluation using a small-scale datacenter. The experimental results demonstrate that our method consistently ensures a lower rate of SLA violation. Compared to existing algorithms, the DRL-based VMP algorithm enables a more balanced utilization of the various resources in the PHs and reduces the total power consumption of the datacenter by more than 10% on average. • Optimizing datacenter efficiency by integrating task scheduling and VM Placement. • Introducing a dynamic task scheduling strategy based on an M/M/1 queue framework. • Employing a deep reinforcement learning (DRL)-based algorithm for efficient VMP. • Our proposed method is validated through a simulation-based evaluation. [ABSTRACT FROM AUTHOR]

As countries worldwide seek to reach net zero emissions within the coming decades, data center sustainability measures are becoming increasingly important as these facilities consume substantial amounts of electricity to support globally expanding digital infrastructure. Yet, local renewable energy production and waste heat utilization are still rarely used in data centers, even though recent high electricity prices have considerably increased the economic value of self-generation and energy efficiency investments. The benefits and cost-effectiveness of data center sustainability measures have also not yet been thoroughly researched in cold-climate regions, where the data center industry is currently growing rapidly. Thus, this paper assesses how solar photovoltaics (PV) and waste heat utilization can effectively be integrated into different cold climate data centers, with a case study analyzing data centers in Finland and northern Japan. The modelled results indicate that solar PV systems can cost-effectively be used to provide renewable electricity to data centers in both Finland and northern Japan, with electricity prices having the largest influence on economic performance. The use of larger external PV systems is shown to be cost-effective at the higher electricity prices seen in 2022, whereas smaller rooftop PV systems and grid electricity provide more economic benefits at 2019 price levels. Selling surplus PV electricity to the grid and using waste heat for district heating is also shown to be effective in data centers located in Finland, whereas battery storage and snow melting are better suited to utilize surplus PV electricity and data center waste heat in northern Japan. [Display omitted] • Solar photovoltaic systems can cost-effectively reduce data center emissions in cold climate countries. • Data center waste heat can provide value in district heating and snow melting applications. • The cost-effectiveness of data center sustainability measures is largely influenced by electricity prices. • The option to sell surplus electricity to the grid incentives larger solar photovoltaic investments in data centers. [ABSTRACT FROM AUTHOR]

For cloud data centers, a performant yet energy-efficient operation is critical for service quality and experience. The growing demand for cloud-based services has led to the development of large-scale hierarchical data center structures, characterized by horizontal expansion and vertical hierarchy, leading to challenges in managing Virtual Machines (VM) at a granular level. The hierarchical arrangement can increase the risk of deployment failures, often stemming from inadequate computational resources on physical hosts, even when the cluster-level resources seem sufficient. While substantial work has gone into managing VMs at the physical host level, there remains a dearth of research under hierarchical data center configurations. To fill the research gap, we address the hierarchy in cloud data centers with a novel two-stage approach named VMM-HHGT, aiming at suppressing VM deployment failures, while balancing the energy consumption and computation resource contention. VMM-HHGT comprises a Hyper-Heuristic-assisted broker (VMM-HH), which can learn the workload patterns and hardware configurations to generate cluster-selection heuristics. An offline training process is incorporated for continuous heuristic evolution with zero overhead on decision-making. Besides, a Game-Theory-assisted hypervisor (GT) is designed for inter-host live VM migration for fine-grained balancing of energy consumption and resource contention. Extensive experiments with traces from real-world VMware data centers show that VMM-HHGT achieves a higher deployment success rate compared to the state-of-the-art approaches, with a well-situated performance in energy consumption and resource contention. • Handling hierarchy in cloud data centers to suppress VM deployment failures. • A two-stage solution for initial VM placement and live migration in the long term. • A Hyper-Heuristic-assisted broker for initial cluster-level VM placement. • A Game-Theory-assisted hypervisor for host-level VM placement and long-term migration. • Effectively suppressed deployment failures with satisfying performance and energy efficiency. [ABSTRACT FROM AUTHOR]

• Chiming in at the advent of 6 G, communication security is increasingly taken into account. • Use efficient iterative method to optimize the variables that affect the cost. • In order to cut system costs down, we propose an interference power price control algorithm based on simulated annealing algorithm, which prioritizes ground users based on this mechanism. • We have innovatively utilized idle roadside units by collaborating with ground users to interfere with eavesdroppers eavesdropping on data, thereby improving the security of the data transmission link. • Ensure that safety performance is maximized in the case of the highest group interest. For the past few years, the security of data transmission links has been increasingly valued. However, with the increasing requirements for the security performance of data transmission links, how to cut the cost of upgrading while improving security has become a serious challenge. In response to the issue of ensuring low-cost and secure data transmission in wireless communication environments with eavesdroppers, this article innovatively starts with idle roadside units and studies how to incentivize roadside units to transmit interference power to interfere with eavesdroppers, thereby efficiently and economically improving the security performance of mobile user data transmission. Firstly, the impact of roadside unit transmission interference power on the security performance of data transmission links was analyzed by establishing a theoretical model; Then, this paper proposes a three-layer model consisting of ground user groups, eavesdropping users, roadside units, and UAVs equipped with mobile edge computing servers. In this model, by encouraging idle roadside units to send interference power, ground users are assisted to safely unload their own data to mobile edge computing servers. Meanwhile, considering the selfishness of idle roadside units, an improved algorithm based on simulated annealing algorithm is proposed to achieve the optimal mapping between ground users and idle roadside units. Finally, through the analysis of simulation results, this scheme can achieve the goal of reducing system costs while improving the security of data transmission. [ABSTRACT FROM AUTHOR]

As the national digitization process accelerates, the number and energy consumption of data centers are increasing year by year. Want to reduce the PUE of the data center. It is correct to start from the perspective of optimizing the energy consumption of air conditioning system operation. In this paper, a predictive optimization control method that is different from previous real-time optimization control is proposed. Predict, optimize, control and adjust the operating parameters of the air conditioning water system in the data center. At the same time, it solves the energy saving problem of the air conditioning system and the lag problem of previous adjustments. Predictive optimization control has broad prospects in research on operation optimization of data center air conditioning water systems. • Reducing air conditioning water system energy consumption in data centers can lower PUE. • Predictive optimization can solve the lag problem of water systems. • After global prediction optimization, energy consumption was saved by 12.9%. [ABSTRACT FROM AUTHOR]

Kernel methods and neural networks (NNs) are two mainstream nonlinear data modeling methods and have been widely applied to industrial process monitoring. However, they both present imperfect properties, so the relevant applications are limited. On the one hand, kernels are not so reconstructable, scalable, and robust to hyperparameters that they suffer performance degradation for large-scale data modeling and monitoring. On the other hand, the high-dimensional parameter space of NNs that is sorted to parameter initialization presents severe anomaly detection performance inconsistency, which makes the industry cautious about using NNs. Motivated by these facts, we propose to integrate kernels and NNs, forming a new model structure that is scalable, reconstructable, and performance-consistent. Specifically, a novel autoencoder-based nonstationary pattern selection kernel (AE-NPSK) is proposed by (1) selecting from the training set the critical edges and interior data as the centers of the radial basis functions in the hidden layers and (2) adaptively adjusting the kernel width in the training procedure. Also, the new NN has strong performance consistency, which facilitates the search for optimal parameters. Finally, we test the performance of the proposed method on the challenging multimode processes. The results validate the efficacy of the proposed method. [ABSTRACT FROM AUTHOR]

In recent years, the energy consumption of data centers (DCs) has shown a sharp upward trend. Given the high investment cost of energy storage, this study introduces the concept of energy sharing within a data center cluster (DCC) and proposes a novel shared energy storage (SES) business model. The model realizes the co-optimization for DCC and SES and is divided into three main layers. Firstly, considering the SES's economy and the DCC's cleanliness, the upper layer establishes a novel multi-objective optimization configuration scheme for SES with lifetime degradation. In addition, a two-stage robust optimization technique for DCC is proposed in the middle layer, considering the demand response mechanism for data loads. Subsequently, the concept of contribution degree (CD) is defined, and a novel benefit distribution mechanism based on cooperative game and CD is presented in the lower layer. The simulation results demonstrate that the proposed business model positively enhances the economic benefits of the SES and DCC. The constructed uncertainty optimization model exhibits strong robustness, and the proposed benefit allocation scheme is fairer and more reasonable. [ABSTRACT FROM AUTHOR]

With the rapid development of global informatization, data center computing tasks are increasing, and the resulting energy consumption problems are particularly prominent in the context of carbon emission reduction, and affect the safe and green operation of data centers. Aiming at the problem of high and large fluctuation energy load of data center servers, this paper presents a method of optimizing data center servers load considering spatial-temporal load distribution of computing resources. Co-optimized allocation of computing resources and load is studied with examples from the property that computing resources is adjustable in time and space dimensions. The results show that the proposed method is able to reduce the peak electric and cooling loads of the data center room by 10.7 %, reduce the total annual electric consumption by 2.66 % and the total annual cooling consumption by 2.46 %, and reduce the economic and environmental costs of energy supply for the data center in the case study, which comprehensively shows that the load optimization of data center servers has been achieved through the optimal allocation of computing resources. [ABSTRACT FROM AUTHOR]

As a critical infrastructure to support the development of digital economy, the scale of data center (DC) power consumption has been growing continuously in recent years. The plug-in of a large number of DCs to the distribution network (DN) presents both opportunities and challenges for the secure, stable, and sustainable operation of the system. In order to address the planning issues arising from the integration of DCs into the DN, this paper proposes a two-stage (planning stage and operation stage) collaborative planning method for DCs and the DN while considering the source-demand uncertainty. Firstly, spatial-temporal demand response characteristic of DCs is explicitly elaborated, and based on the energy consumption model, a two-stage stochastic planning model considering conditional value at risk (CVaR) for DCs and the DN is constructed using probabilistic stochastic simulation, which obeys low-carbon and security, and satisfies the source-network-load constraints of the system. The siting and sizing for DCs and renewable generation, feeder expansion for the DN are determined to minimize the investment cost during the planning stage, while the power output for the devices considering uncertainties from load and renewable energy are optimized to minimize the operation cost during the operation stage. Secondly, the L-shaped decomposition algorithm is proposed to solve the two-stage stochastic optimization problem by reconstructing the problem into a master problem and subproblems, where the dual multiplies are calculated to generate feasibility cuts and optimality cuts to achieve iterative solving. Finally, numerical case studies on a 33-bus distribution system demonstrate that the participation of DCs in demand response can achieve a significant reduction in the investment and operation cost of the DN, and further promote the consumption of distributed renewable energy sources. • The spatial-temporal characteristics of DC demand response is explicitly elaborated and the mathematical formulation is proposed. • A two-stage stochastic collaborative planning approach of DC and DN based on CVaR is proposed. • Guidance for investment decisions for DSO with large numbers of DCs plug-in under uncertainties. • Demand response of DCs can effectively support renewable energy consumption and reduce the cost of DN. [ABSTRACT FROM AUTHOR]

As the demand for Internet services continues to rise, large-scale Internet companies have expanded their data center infrastructure across multiple global regions. The inherent spatial-temporal flexibility in scheduling workloads allows data center clusters to dynamically adjust their power consumption without overloading the transmission network. In this work, we introduce a novel approach for treating a geo-distributed data center cluster as a non-wire alternative resource and explore its potential role in the energy balancing market. Our methodology manages a geographically distributed data center cluster as a balance-responsible entity by employing optimal workload scheduling that considers its inherent spatial-temporal flexibility and market arbitrage opportunities. To achieve this, we develop a two-layer stochastic receding horizon control optimization algorithm to formulate a strategic bidding strategy for the data center cluster. To validate the effectiveness of our proposed approach, we conduct numerical experiments using real-world data center production traces and data from the Nordic balancing market, demonstrating its practical applicability and performance. • Explored spatial-temporal flexibility of data center cluster as non-wire alternative. • Applied a two-layer SRHC algorithm for online biddings and workload schedules. • Applied real-world data traces to illustrate the effectiveness and accuracy. • Real-world data showed the scheme's impact on cost reduction in balancing markets. [ABSTRACT FROM AUTHOR]

In the age of digitalization, data centers are critical facilities that help people and businesses in many sectors. Data centers, like other assets requiring high investment, face several risks during their operation. Also, data centers seek sustainable sources to boost their operations when selecting the most appropriate location. Therefore, the location of a data center is chosen based on sustainability and risk assessment. To the best of our knowledge, no study has been conducted for data center risk evaluation with environmental and sustainability assessment. This paper evaluates the eight main criteria and the forty-five sub-criteria for selecting the most appropriate location for the data center when considering the fifty different economies in different business regions. For this particular purpose, the continuous intuitionistic fuzzy sets (CIFS)-based analytical hierarchy process (AHP) is presented to define the weights of each criterion and sub-criterion. Then, the CIFS-based technique for order of preference by similarity to the ideal solution (TOPSIS) is developed to rank the fifty economies integrated with the weights from the CIFS-based AHP. Moreover, eight different scenarios were conducted, and comparison studies were presented. Finally, Denmark, Switzerland, and The Netherlands are the top three selections based on the criteria. [Display omitted] • Fuzzy multi-criteria methods are developed to evaluate data center locations. • Natural, energy, and infrastructure risks are the top three pillars. • Economies are ranked based on the eight scenarios with sustainability and risks. • Denmark tops the charts with the highest score associated with fifty economies. • Based on each income level, Denmark, Brazil, and Tunisia are suitable locations. [ABSTRACT FROM AUTHOR]

With the rapid expansion of computational infrastructure, the electricity consumption of geographically dispersed Internet Data Centers (IDCs) has imposed a substantial influence on the power grid. Geographically distributed IDCs, which present a versatile electricity demand with significant operational flexibility, hold potential for mitigating cost issues and balancing regional supply and load demand. To exploit this flexibility, this paper has developed a dual-drive coordinated operational strategy, which integrates locational electricity prices and unit carbon emission costs based on the dynamic carbon emission intensity (DCEI) of various regions as interaction signals to improve coordination effectiveness. The upper-level determines the consumption behaviour of IDCs, the lower-level provides feedback on dual-drive signals factoring in IDCs' involvement in grid operation, whilst an iterative algorithm is suggested to solve the bi-level model. Case studies based on real-world datasets demonstrate that the proposed operation strategy can achieve arbitrage through multi-regional carbon-electricity markets, significantly reduce operating costs, and support the rapid development of IDCs. Additionally, this strategy facilitates optimal matching of power load in IDCs with renewable energy sources, thereby mitigating imbalances in regional renewable energy supply and demand. • Analysing the necessity of integrating IDCs into the carbon market. • Proposing a dual-drive coordination to activate spatiotemporal flexibility of IDCs. • An iterative algorithm for solving the bi-level model is designed. • Comparing impacts of carbon policies in IDC costs and renewable energy accommodation. [ABSTRACT FROM AUTHOR]

Tobacco drying is a highly energy-consuming process that requires a large amount of thermal energy. In this study, the waste heat from a data center is utilized to reduce fossil energy consumption in the tobacco drying process. A heat pump system is designed to address the issue of the waste heat temperature from the data center not meeting the required temperature for tobacco drying. Multi-objective optimization is performed with the goal of maximizing the specific moisture extraction rate and minimizing the specific investment cost, and seven hydrofluoroolefins are selected as the circulating working fluids. Coal-fired tobacco drying and Air Source Heat Pump Tobacco Drying (ASHPTD) are selected as comparison benchmarks to evaluate the energy efficiency and environmental benefits of the Data Center Waste Heat Tobacco Drying (DCWHTD) method in this study. The results indicate that R1234ze(Z) is the best working fluid. The specific moisture extraction rate of DCWHTD is improved by 36.69% and 20.21% compared to the coal-fired tobacco drying and ASHPTD, respectively, demonstrating enhanced energy efficiency. Although the specific investment cost of DCWHTD is higher compared with the coal-fired tobacco drying, it is 21.00% lower than that of ASHPTD. DCWHTD produces fewer emissions, the life-cycle carbon emissions are 26.93% and 16.80% lower than that of the coal-fired tobacco drying and ASHPTD, respectively. Furthermore, waste heat recovery improves the energy reuse efficiency of the data center by 64.92%. The tobacco drying method proposed in this study has better comprehensive performance and greater application potential. • Waste heat of data center is recovered by heat pump and used for tobacco drying. • Optimal design of system is obtained by multi-objective optimization method. • R1234ze(Z) is determined to be the optimal working fluid among the selected fluids. • Economic and environmental advantages over common drying methods are revealed. • Energy efficiency improvement of data center by waste heat recovery is evaluated. [ABSTRACT FROM AUTHOR]

Convolution is a fundamental and computationally expensive primitive and finds ubiquitous in deep neural networks (DNNs). The evolving DNNs have spurred the emergence of numerous accelerators and they successfully achieve high throughput. However, for DNN inference with small batch sizes, the computational resources of the accelerators are often under-utilized, and the overhead of offloading is significant. Compared to accelerators, the CPU can better meet fast response requirements of inference, flexibly handle various models, and is suitable for various scenarios (from edge to data center). Therefore, CPU remains an attractive platform for DNN inference, despite the sub-optimal performance, and resource efficiency. In this paper, we propose CUTE, a scalable CPU-centric and ultra-utilized tensor engine for convolutions. It co-designs data flow and hardware architecture to leverage the data reuse and parallelism of convolutions. CUTE is composed of several small tensor elements (TEs) and two-level buffers. It employs a decoupled access-execution architecture and greedy strategy to feed data to TEs, enabling it to achieve ultra utilization and great scalability. CUTE is tightly coupled with the CPU to minimize offloading latency, thereby providing efficient convolution computing capabilities for the system. Experimental results show that under the same bandwidth, CUTE achieves an average performance improvement of 3.8x compared with the CPU AVX512 unit and 1.6x compared with the CPU AMX unit. Besides, CUTE achieves a speedup of 7.0x and 3.9x over Nvidia V100 GPU and Eyeriss accelerator respectively, due to higher utilization of computing units. [ABSTRACT FROM AUTHOR]

Recently, more and more global attention has been paid to data centers' energy consumption and climate impact. Serving as the backbone of the digital world, the energy and carbon-intensive data centers have to find a green expansion way to fulfill their obligations in building a carbon-neutral society. Therefore, this paper proposes a net-zero carbon operation oriented data center expansion scheme for the central government to formulate incentive policies. The economic expansion model minimizes newly established data centers' investment and electricity consumption costs. Geographical location, on-site renewable utilization, and green certificate purchase are integrated and considered to achieve net-zero carbon operation by reducing total electricity consumption, increasing green power consumption, and offsetting inevitable carbon emissions. A case study in China is conducted, and results under different geographical location scenarios, power supply mix scenarios, and green certificates market scenarios are analyzed and compared. It is found that the zero-carbon operation of new data centers mainly relies on clean main-grid power, but a more decentralized layout and a provincial green certificate market can increase the contribution of on-site renewable energy. Lastly, four detailed policy implications for stimulating China's data center expansion are discussed. • A data center expansion scheme considering net-zero carbon operation is proposed. • Expansion model minimizes newly established data centers' expansion costs is built. • Three measures are integrated to achieve data centers' net-zero carbon operation. • Case study in China is conducted and core findings and key implications proposed. [ABSTRACT FROM AUTHOR]

Boiling two-phase transition is considered to be the ultimate effective cooling strategy for the next generation of high-power devices in the near future in various application scenarios, such as power electronics, radar systems, and computer data centers. The liquid-solid interface plays a crucial role in the boiling phase change and the bubble evolvement. This study proposes a novel architecture of micro-nano structures using a unique hybrid dividing zone array, which can not only enhance the onset of nucleate boiling (ONB) and dramatically elevate the heat transfer coefficient (HTC), but also accelerate liquid replenishment, pushing the critical heat flux (CHF) to a much higher level. This is achieved through a superhydrophilic liquid storage zone that provides timely and continuous liquid supplementation for boiling phase change during the drying out period of the most copper surface. Compared to the plain surface copper sample, the proposed hybrid structured sample achieves about 5–7 °C earlier of ONB, up to 293% enhancement of the maximum HTC, and up to 242% improvement of CHF. Overall, this study demonstrates an effective surface enhancement approach that significantly and comprehensively improve boiling heat transfer performance, making it highly promising for cooling high-power applications. • Hybrid micro-nanostructures were constructed using a unique hybrid dividing zone array. • 5–7 °C earlier of onset of nucleate boiling was achieved. • Up to 293% enhancement of the maximum heat transfer coefficient was accomplished. • Up to 242% improvement of critical heat flux was realized. • Stable and continuous boiling was sustained when large bubble fully covering heating surface. [ABSTRACT FROM AUTHOR]

Data centers consume large amounts of electricity and water specifically for cooling purposes. Integrating free cooling technologies is an essential part of broad efforts to reduce data centers' energy, carbon, and water footprints. Adding a new technique to the existing portfolio of free cooling technologies for data centers, we present how all-day passive radiative cooling can reduce energy and water consumptions of data centers. All-day passive radiative cooling is a free cooling technique, in which sky-facing surfaces with high solar reflectivity and midinfrared emissivity spontaneously reject heat into the sky, without consuming electricity and evaporating water. Using custom-built physics-based models, we simulate three configurations that integrate all-day radiative free cooling into data centers. We show that data center cooling energy and water consumptions differ substantially depending on the integration configuration. Particularly, direct free cooling of data center return air gives the highest energy saving with an annual average of around 20.0% under the tropical climate of Singapore which hosts a large number of data centers, whereas free cooling of compressed refrigerant prior to the chiller condenser gives the highest water saving with an annual average of around 84.0%. These results offer multiple viable options for integrating radiative free cooling into data center cooling systems to simultaneously achieve water and energy savings. • Radiative free cooling is added to the existing portfolio of free cooling technologies for data centers. • Detailed physics-based models of a data center cooling system were built. • The effects of weather and operation conditions on the cooling system performance were investigated. • Direct free cooling of data hall return air yields the most energy saving with an annual average of around 20.0%. • Free cooling of compressed refrigerant of the chiller cycle gives the most water saving with an annual average of 84.0%. [ABSTRACT FROM AUTHOR]

In order to develop the green data center driven by solar energy, a solar photovoltaic (PV) system with the combination of compressed air energy storage (CAES) is proposed to provide electricity for the data center. During the day, the excess energy produced by PV is stored by CAES. During the night, CAES supplies power to the data center, so as to reduce the cost of peak electricity consumption at night. To evaluate the system performances, thermodynamic and economic models are established. Thereafter, system performances under design conditions and the effects of system parameters are analyzed. The results indicate that under design conditions, for the 17.5 MW data center, the required solar PV area is 257075 m2, and the highest PV power can reach up to 55 MW. The all-day efficiency of the PV system is 18.37 %. In this situation, a CAES system with 17.5 MW × 7.88 h is configured, and the variations of CAES parameters with time are obtained in a day. The round-trip efficiency is 64.88 % and the energy storage density is 5.02 kW·h·m−3. The total exergy destruction of the whole system within 24 h can be up to 1581001 kW h. For the economic performance, the total cost is $93.87 M and the payback period is 11.84 years. • A novel solar photovoltaic-compressed air energy storage system is proposed. • The parameters of air storage reach a steady state after 30 days of operation. • The models of thermal-economic performances are established. • The system parameters are analyzed. [ABSTRACT FROM AUTHOR]

Aiming at the carbon peaking and carbon neutrality goals, the water resource and energy management of internet data center involves many aspects, including the interaction among internet data center, power grid, water network and heat consumer. As an important demand response resource for the power grid, the interaction between both can improve the flexibility of the power grid, and ensure the power supply reliability and operation economy of internet data center. Existing references assume that regional integrated energy system can obtain all data information and scheduling permissions of internet data center, ignore the autonomous decision-making ability of both, resulting in its application has some constraints. The operation pattern switchover of internet data center and data allocation among multiple internet data center s are ignored, it is not possible to fully play the flexibility characteristics of internet data center. The study of interaction between internet data center and power grid simplifies the power grid as an ideal power source, ignoring its operational constraints. Internet data center configured with renewable energy, electric energy storage device and water-cooling system is taken as one of the research objects in this paper, and regional integrated energy system formed by the coupling of distribution network and water network as the other. The water-energy coupling and cooperative operation between above two are studied for solving the above problem. The proposed method has been validated in the demonstration project of the "East Data and West Computing" big data industrial park in Qingyang, Gansu, China. The results of example show that water-energy demand response and waste heat reuse can help to reduce the operational cost of internet data center, improve scheduling flexibility of regional integrated energy system, and enhance the capacity of renewable energy utilization. The operational cost of internet data center and regional integrated energy system is sensitive to emergency capacity of electric energy storage device, cost coefficient of the power demand response of electric energy storage device and water demand response of reservoir, data load flexible scheduling strategy, which is consistent with the operational status of the demonstration project. • Refined model of regional integrated energy system and internet data center is built. • Water-energy coupling model is established considering data load and waste heat reuse. • Water-energy demand response model of internet data center is proposed. • Three-stage collaborative schedule model is proposed for data privacy and security. [ABSTRACT FROM AUTHOR]

• Performance evaluation of three different cooling methods for data centers is conducted. • Single-phase immersion cooling has advantages in heat transfer and energy consumption. • Correlation between oil properties and immersion cooling performance is established. • Dynamic viscosity has the greatest impact on the flow and thermal performance of oil coolants. This paper conducts a comprehensive performance evaluation and optimization of single-phase immersion cooling (SPIC) systems to address challenges in data center server cooling. A three-dimensional model of blade servers with different cooling methods is constructed and numerically solved to compare thermal transport performance, flow characteristics, and energy consumption performance. Furthermore, the influence mechanisms of different oil coolants and operating conditions on SPIC system performance are studied. The results indicate that the SPIC scheme exhibits significant advantages in terms of heat transfer capability, heat transfer uniformity, and flow resistance loss. Moreover, the power usage effectiveness (PUE) of SPIC systems decreases by 20.8 % and 17.6 % compared to forced air cooling and water cooling plate solutions, respectively. The dynamic viscosity of oil coolants has the greatest impact on the SPIC performance. Moreover, high-viscosity silicone oil increases the flow resistance of SPIC systems and weakens the convective heat transfer capability between the coolant and components. By ensuring safe operation, increasing the inlet temperature of oil coolants reduces flow resistance losses and enhances heat transfer capacity. Additionally, initially increasing the mass flow rate improves convective heat transfer capacity. However, the heat transfer improvement gradually diminished beyond a certain threshold while flow resistance losses increased significantly. [ABSTRACT FROM AUTHOR]

First report of a unique microporous Ho-based metal–organic framework which changes its physical color from white to pink upon changing incident light sources. [Display omitted] • A novel zwitterionic Holmium-based microporous metal-organic framework, H1 , was designed. • The MOF H1 shows unique optical properties as evidenced by distinct color changes under different lighting conditions. • A survey using the Cambridge Crystallographic Data Centre demonstrates that MOF H1 is the first reported example of this unique optical feature. Lanthanide metal–organic frameworks (Ln-MOFs) have gained significant attention in recent years due to their diverse characteristics ranging from material science to biomedical applications. Although Ln-MOFs possess properties that are common to all MOF categories, they can exhibit unique structural combinations owing to their high coordination numbers and distinctive optical properties. Herein, we report the Ho(III)-based zwitterionic MOF H1 which crystallizes in the monoclinic space group C2/c. The nine-coordinate Ho(III) ion presents a configuration related to the gek2 topology. Remarkably, H1 features unique optical properties, namely, the physical change of color from white to pink upon changing the incident light source from sunlight to fluorescent light. A literature review through the Cambridge crystallographic data center (CCDC) reveals this being the first report of a Ho(III)-MOF showing such optical behavior. [ABSTRACT FROM AUTHOR]

While providing efficient and convenient cloud services, data center also brings great pressure to energy consumption and environment. The rise of server temperature not only increases the refrigeration cost, but also seriously affects the operation safety of the data center. Effective analysis and prediction of data center temperature is not only conducive to preventing server overheating and shutdown, but also crucial to data center task scheduling, resource allocation optimization and energy efficiency improvement of data center. Therefore, this article proposes a Two-stage Gated Recurrent Unit (GRU) temperature prediction algorithm with self-healing mechanism. The algorithm establishes a prediction model for the important parameters affecting temperature prediction - CPU utilization, and takes the output of the model as the input parameter of the server temperature prediction model, which fits the changes of each parameter more accurate. To avoid the decrease in prediction accuracy caused by new operating conditions that have not been learned before and changes in physical environmental factors during the operation of the model, a self-healing mechanism is proposed to ensure the prediction accuracy of the model. Experiments show that our prediction model can accurately predict the inlet temperature evolution of the server with dynamic workload. It reduces the prediction error (RSME) to 0.280, and the average prediction temperature difference is only 0.675, which is 10 % higher than the single stage prediction accuracy. The use of Two-stage prediction methods in other machine learning methods can also improve prediction accuracy. Based on the prediction model, this paper proposes a task scheduling algorithm that minimizes temperature difference. The algorithm can make the temperature between servers more balanced after task allocation, effectively reducing the number of servers running at high and low temperatures in the data center, avoiding refrigeration waste, and achieving energy conservation in the data center. [ABSTRACT FROM AUTHOR]

A more comprehensive understanding of the impact of artificial intelligence (AI) on energy transition and carbon emissions could help to use AI to achieve carbon neutrality. To this end, the STIRPAT approach, the mediation effect technique and the panel threshold technique are developed using the panel data in 69 countries from 1993 to 2019. The results show that: (i) AI promote energy transition and carbon emission reduction, and trade openness (indicated by imports, exports and total trade volume) has the mediating effect. (ii) There is a single-threshold of trade openness in the impact of AI on carbon emissions. When trade openness is below the threshold, AI has an insignificant impact on carbon emissions; when trade openness crosses the threshold, AI has a significant negative impact on carbon emissions. There is a double-threshold of trade openness in the impact of AI on energy transition. When trade openness is lower than the first threshold, the impact of AI on energy transition is not significant; When trade openness is higher than the second threshold, the positive impact of AI on energy transition is increased. (iii) When considering the heterogeneity of income levels and AI levels, the trade threshold for achieving carbon emission reductions in the high-income group is lower than that of the global group, and the trade threshold for achieving carbon emission reductions in the low-AI level group is higher than that of the global group. While this study unequivocally delineates the affirmative role of artificial intelligence in carbon emission reduction and energy transformation, particularly in the context of trade openness, we concurrently acknowledge that this viewpoint is not devoid of contention. Amidst the rapid advancement of technology and the landscape of open trade, we discern the presence of counterarguments. The efficacy of artificial intelligence is susceptible to the influence of multifaceted factors. It is imperative to consider associated factors, such as the significant energy consumption required for storing and cooling data centers and servers. The study's conclusions aid policymakers in devising nuanced emission reduction policies tailored to specific needs. [Display omitted] • Investigate the effect of artificial intelligence on carbon emissions and the energy transition. • Trade openness is incorporated in the research framework of artificial intelligence, carbon emissions, and energy transition. • Artificial intelligence can promote carbon emission reduction and energy transition. • Trade openness has mediating effects and threshold effects. • Heterogeneity across income levels and artificial intelligence development levels is investigated. [ABSTRACT FROM AUTHOR]

This research work introduces a novel approach to enhance the performance of distributed data warehouses. Distribution of data has been done across multiple data center for the underlying data warehouse to ensure parallel processing that improves the performance of the system. The proposed system focuses on data availability locally, resulting in faster query processing times. In order to further optimize the query processing and to alleviate the centralized data warehouse load, the study suggests distributing the cuboids at different sites to maintain the lattice of cuboids logically. Algorithms have been proposed to execute OLAP aggregation operations that significantly enhanced query response times in distributed data warehouse. Empirical tests on real-world datasets proved the method's effectiveness, showcasing its superiority in terms of reduced OLAP query execution time, enhanced space efficiency, minimized network transmission delays, reduced bandwidth requirements and more predictable and stable data transmission compared to existing approaches. • Novel approach optimizes distributed data warehouses. • Queries improved via distributed cuboid placement, retaining lattice structures. • Real-world tests validate faster queries, reduced network delays, stability. • Proposed algorithm makes 23% reduction in OLAP query time in a case study. • Peak Transmission Rate increased by 10 times than existing method. [ABSTRACT FROM AUTHOR]

Cloud computing is one of the powerful engines driving global industrial upgrading and the booming digital economy. However, the explosive growth of cloud data centers (DCs) has resulted in inevitable energy consumption and carbon emission problems. Therefore, constructing energy-efficient and sustainable DCs will be essential for green cloud computing. This review makes several efforts to thoroughly investigate and track the research progress and routes to sustainable DCs. Firstly, we construct a new conceptual model of sustainable DCs to cover cutting-edge research results and indicate future evolutionary directions. Secondly, this review provides a comprehensive survey of five topics from a technical perspective: workload management, virtual resource management, energy management, thermal management, and waste heat recovery. Subsequently, some real-world datasets relevant to the topics, including workload traces, renewable energy data, and electricity price traces, have been specifically collected to support researchers in conducting further research. Finally, based on observations of existing works, we highlight some salient technical challenges and promising solutions to provide sensible energy and carbon reduction suggestions in sustainable DCs. • This review presents research advances and routes for sustainable data centers. • A new conceptual model of data centers is constructed to cover cutting-edge topics. • This review categorizes and analyzes five hot topics from a technical perspective. • Some real-world datasets are collected to support related research. [ABSTRACT FROM AUTHOR]