Computing Data Centers Research Articles

Cross regional online food delivery: Service quality optimization and real-time order assignment

Online food delivery (OFD) represents a rapidly evolving e-business application that leverages cloud computing data centers, playing a crucial role in meeting the demands of urban lifestyles. With diverse order fulfillment features and increasing expectations for service quality, the task of effectively assigning riders for timely long-distance, cross-regional deliveries presents a significant engineering challenge. Previous studies often relied on traditional rider allocation methods that fail to account for varying capacities, or they utilized non-intelligent systems that did not adequately address fluctuating order demands and service delays. In this study, we introduce a robust Mixed Integer Linear Programming (MILP) optimization framework designed to minimize the total service time and delivery cost for cross-regional orders. This framework divides a large OFD area into multiple regions and utilizes both transfer vehicles and riders to optimize deliveries. To enhance the predictive accuracy of our model, we incorporate advanced machine learning techniques. Specifically, we employ the Long Short-Term Memory (LSTM) model to forecast regional order demands accurately, reflecting the dynamic nature of the marketplace. Additionally, Extreme Gradient Boosting (XGBoost) is tailored to dynamically predict travel times from restaurants to customer locations, facilitating more precise scheduling and resource allocation within the MILP framework. These machine learning techniques significantly bolster the MILP framework by providing detailed, accurate predictions that improve decision-making processes and adaptability to real-time conditions. Acknowledging the complexity of this optimization problem, we further enhance our approach by integrating a meta-heuristic algorithm, Adaptive Large Neighbor Search (ALNS), which efficiently assigns orders to the appropriate transfer vehicles and riders within polynomial time. Our Cross Regional Online Food Delivery (XROFD) system is meticulously designed to optimize both customer satisfaction and rider incentives. Simulation experiments confirm that the XROFD system not only reduces service times and delivery costs but also markedly enhances customer satisfaction and provides superior incentives for riders, outperforming existing state-of-the-art methods.

V2X Communication Computing Resource Collaboration Technology Based on Deep Reinforcement Learning

In traditional cloud computing, data needs to be transferred from distributed data sources to remote cloud data centers for high-performance computing. However, tasks often experience network transmission delays of hundreds of milliseconds or more from data sources to cloud data centers, which does not meet the need for low latency in V2X (Vehicle to Everything) systems. On the basis of the existing heterogeneous task scenario, this paper proposes a communication computing resource collaboration technology scheme based on heterogeneous task, which can meet the requirements of delay and cost when facing tasks with different delay requirements and different divisibility. This scheme uses the communication computing resources of the roadside unit and the service vehicle, so that the computing task of the task vehicle can be completed within the delay requirement and the cost is minimal. We use the deep reinforcement learning method to study, and finally the performance of the proposed scheme is verified by simulation.

An expandable and cost-effective data center network

With the rapid growth of data volume, the escalating complexity of data businesses, and the increasing reliance on the Internet for daily life and production, the scale of data centers is constantly expanding. The data center network (DCN) is a bridge connecting large-scale servers in data centers for large-scale distributed computing. How to build a DCN structure that is flexible and cost-effective, while maintaining its topological properties unchanged during network expansion has become a challenging issue. In this paper, we propose an expandable and cost-effective DCN, namely HHCube, which is based on the half hypercube structure. Further, we analyze some characteristics of HHCube, including connectivity, diameter, and bisection bandwidth of the HHCube. We also design an efficient algorithm to find the shortest path between any two distinct nodes and present a fault-tolerant routing scheme to obtain a fault-tolerant path between any two distinct fault-free nodes in HHCube. Meanwhile, we present two local diagnosis algorithms to determine the status of nodes in HHCube under the PMC model and MM* model, respectively. Our results demonstrate that despite the presence of up to 25% faulty nodes in HHCube, both algorithms achieve a correct diagnosis rate exceeding 90%. Finally, we compare HHCube with state-of-the-art DCNs including Fat-Tree, DCell, BCube, Ficonn, and HSDC, and the experimental results indicate that the HHCube is an excellent candidate for constructing expandable and cost-effective DCNs.

Quantum Enhanced Josephson Junction Field-Effect Transistors for Logic Applications

The need of data centers for cloud and network computing has dramatically increased power consumption. In 2014, data centers in the U.S. consumed an estimated 70 billion kWh, representing about 1.8% of total U.S. electricity consumption [“United States Data Center Energy Usage Report," published in June 2016 and supported by the Federal Energy Management Program of the U.S. Department of Energy]. Worldwide, the International Energy Agency estimates that currently about 1% of all global electricity is used by data centers. It is expected that by 2030 the electricity use of information and communications technology may exceed 20% of all global electricity. The microelectronics industry has been focusing on aggressively shrinking the size of logic and memory devices to reduce power consumption. However, this scaling is going to eventually stop as we approach the fundamental materials limit. Moreover, the power dissipation due to the increased number of transistors is also expected to be a limiting factor for processor performance.Superconducting computing has been suggested as a promising approach for low-energy, power-efficient circuit applications. Moreover, using superconducting interconnects can further reduce power consumption. Josephson junction field-effect transistors (JJFET, Fig. a) have recently emerged as a promising candidate for superconducting computing. JJFETs are particularly useful for low power consumption applications as they are operated with, in the superconducting regime, zero voltage drop across its source and drain. Feasibility of using JJFETs for logic operations has been explored [1]. Moreover, high noise margin has been demonstrated for the logic inputs [2]. Compared to single flux quantum approach, JJFET circuits can use designs similar to conventional silicon MOSFETs [1].For JJFETs to perform logic operations, the gain-factor (αR) value must be larger than 1. Here αR = dIc/d(Vg–Vt) x πΔ/Ic, Δ is the superconducting gap, Vg the gate bias voltage, Vt the threshold voltage. Ic ∝ exp(-L/ξc) is the critical supercurrent, where L is the channel length and ξc the carrier coherence length [3]. In a conventional JJFET, ξc ∝ (Vg–Vt)0.5 (Fig. b), and thus dIc/d(Vg–Vt) is small (Fig. c). This translates to a requirement of superconducting transition temperature of ~ 400K for αR larger than 1, far exceeding any recorded critical temperatures. As such, it is impossible to use conventional JJFETs for logic operations.Here, we propose a novel type of JJFET based on quantum phase transition, such as the excitonic insulator (EI) transition in a type-II InAs/GaSb heterostructure [4,5], for low-energy, power-efficient logic applications. The nature of the collective phenomenon in the EI quantum phase transition can provide a sharp transition of the supercurrent states (e.g., ξc ∝ (Vg–Vt)5 and dIc/d(Vg–Vt) very large, as shown in Figs. b and c, respectively) which will enable αR larger than 1 with an easy-to-achieve superconducting transition temperature of ~ 40K. In this talk, we will present some preliminary results demonstrating that indeed the gain factor in these quantum enhanced JJFETs can be greatly improved, thus making them a promising candidate for logic applications.Sandia National Laboratories is a multimission laboratory managed and operated by NTESS LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. DOE’s NNSA under contract DE-NA0003525. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. DOE or the United States Government.

Evolving High-Performance Computing Data Centers with Kubernetes, Performance Analysis, and Dynamic Workload Placement Based on Machine Learning Scheduling

Evolving High-Performance Computing Data Centers with Kubernetes, Performance Analysis, and Dynamic Workload Placement Based on Machine Learning Scheduling

Synergies and Challenges in the Integration of Cloud Computing and Deep Learning: Current Status, Interconnectedness, and Future Directions

This article reviews the status and recent developments in the integration of cloud computing and deep learning, as well as the interrelationship between these two technologies. The paper explores the intersection of cloud computing and deep learning in addressing cybersecurity challenges. Amidst the rapid expansion of the worldwide public cloud services market, the vulnerability to cyber-attacks and breaches in data management is on the rise. Different intrusion detection systems use different deep learning techniques to improve the effectiveness of intrusion detection in cloud computing environments. Additionally, the use of encryption technology and the corresponding deep learning retrieval technology further improves the security of cloud data. Moreover, the paper deeply studies how the scheduling mechanism of deep reinforcement learning can optimize the performance of cloud services by efficiently allocating resources and solving the problem of slow cloud service speed. It also derives the optimal energy strategy through deep neural networks to address the energy consumption challenges in cloud computing data centers. This article also reviews the five emerging architectures of cloud computing and explores the role of deep learning within these frameworks. Finally, it analyzes some of the challenges facing the future of cloud computing and deep learning, including the security and confidentiality of cloud computing, as well as low latency and high throughput optimization in the field of deep learning. In summary, this article provides insight into current trends, challenges, and future prospects for the evolving integration between cloud computing and deep learning.

District heating utilizing waste heat of a data center: High-temperature heat pumps

Data centers are energy-intensive facilities with substantial low-grade waste heat. High-temperature heat pumps can be critical in boosting the data center’s waste heat for district heating, improving the system-level energy efficiency of data centers, and reducing CO2 emissions in district heating. This study built thermodynamic models to assess high-temperature heat pumps with six configurations using low global warming potential refrigerants to supply heat up to 120 °C. The heat pump configurations include single-stage or two-stage cycles with advanced components, such as internal heat exchanger, economizer, flash tank, or parallel compressor. The refrigerants include R1234ze(Z), R1233ed(E), R1224yd(Z), R600, and R600a, and R245fa is used as a reference. A case study was carried out to recover the waste heat from the Frontier high-performance computing data center and provide hot water for district heating at the US Department of Energy’s Oak Ridge National Laboratory campus. The optimized performance of high-temperature heat pumps is characterized with various effectiveness of internal heat exchangers, and the operating parameters of economizer or flash tank, as well as their combination. The results show that the configurations of two-stage cycles with internal heat exchanger + flash tank and internal heat exchanger + economizer/parallel-compressor provide the highest coefficient of performance under scenarios of the maximum allowable value and a fixed value (0.3) of the internal heat exchangers’ effectiveness, respectively. R1234ze(Z) and R600a are the most promising refrigerants, considering trade-offs between the coefficient of performance and the volumetric heating capacity. The single-stage cycle with internal heat exchanger + economizer/parallel-compressor using R1234ze(Z) is recommended for utilizing Fronter’s waste heat in district heating. A one mega-watt high-temperature heat pump will reduce 33,100–33,200 metric tons of CO2 emission annually, corresponding to 85.4 %–85.6 % of equivalent CO2 emissions from natural gas boilers. This study provides good guidelines for designing and deploying high-temperature heat pumps to support sustainable data centers and decarbonize district heating in the US.

The butterfly effect of cloud computing on the low-carbon economy

This paper investigates the long-term butterfly effect of cloud computing on the low-carbon economy against the dual backdrop of the digital age and carbon neutrality. Utilizing a full-sample and sub-sample approach, the study identifies the complex interrelationship between the China Cloud Computing Index (CCI) and the Low Carbon Index (LCI). The quantitative examination reveals how CCI impacts LCI – both favorably and unfavorably. The positive effects suggest that cloud computing serves as a motivator in the environmentally friendly economy. In case of negative impacts, however, it is not possible to always determine the incentive effects owing to the large amounts of energy consumption generated by cloud computing data centers. Meanwhile, the positive influence of LCI on CCI indicates how the pursuit of low carbon-economy (as a goal) will bring opportunities for the explosive growth of the cloud computing industry. Cloud computing has increasingly prompted the fluctuation of carbon emissions in the whole ecosystem. By exploring this butterfly effect, this study digs deeper into the complex interrelationship between the discussed indexes. It is expected that the study outcomes will offer meaningful recommendations for the two main vectors of economic development - greenization and digitalization.

Optimizing Task Scheduling in Cloud Data Centres with Dynamic Resource Allocation Using Genetic Algorithm (TSOGA)

Nowadays, Massive business applications are increasingly giving attention to cloud computing data centres because of its high potential, adaptability, and efficiency in supplying several sources of both software and hardware to support networked consumers. The criteria for autonomy of virtual machines necessitate a flexible resource allocation strategy for Virtual Machines (VMs) .The majority of resource utilization models were inaccurate, making it impossible to determine the virtual machine's energy usage directly from the hardware. Due to the size of modern data centres and the constantly changing character of their resource supply, efficient scheduling solutions must be developed to oversee these resources and meet the objectives of both cloud service providers and cloud customers. Hence an algorithm called Task Scheduling Optimization based Genetic Algorithm (TSOGA) has been proposed to dynamically allocate the resources in pursuit of scheduling the tasks in cloud data centers. The proposed module initially focuses on task scheduling process, followed by optimized running time of task execution. For data centres with dynamic resource allocation, the goal of TSOGA is to efficiently assign jobs to resources while minimizing execution time and optimizing resource utilization. Thus, to manage the data centres while achieving high levels of efficiency in resource allocation, we constructed a virtual node for our research. Incorporation of Genetic Algorithm is to determine an ideal or nearly ideal schedule for carrying out tasks using the available resources while taking into account a variety of restrictions and goals, such as minimizing execution and waiting time of task during dynamic scheduling process and efficient resource utilization.

Bi-objective optimization and reliability analysis in unreliable M/G/1 retrial queues with rational customers, reserved idle times and setup times

ABSTRACT Energy consumption arising from no-loads and failures is a common and challenging problem in the design of service systems. To analyze the impact of no-load consumption and failure consumption on the energy consumption simultaneously, this paper considers an M/G/1 retrial queue with reserved idle times, setup times and server breakdowns. First, the steady-state condition and performance measures of the system are obtained by the embedded Markov chain and the supplementary variable method. Then, the reliability indicators of the system are analyzed, and an algorithm for solving the system reliability R ( t ) is constructed based on the Monte Carlo method. In order to better optimize the system, the customer’s equilibrium joining strategy is derived, and the socially optimal joining strategy is obtained by the particle swarm optimization (PSO) algorithm. From the decision-maker viewpoint, a bi-objective optimization model is constructed to minimize the expected energy consumption and the expected waiting time of retrial customers, and solved by the non-dominated ranking genetic algorithm II (NSGA-II). Finally, we illustrate an application of the proposed analytical approach in the cloud computing data center, which makes it possible to achieve an appropriate balance between energy consumption and quality of service (QoS).

Application of SDN-IP hybrid network multicast architecture in Commercial Aerospace Data Center

Abstract The increasing amount of massive data generated by commercial spacecraft in orbit puts forward higher and higher requirements for the stability, reliability, and computing power of computer systems for commercial aerospace data center. Data center computer systems are gradually transforming from X86 architecture and IP network model to a platform model with cloud computing and software-defined network (SDN) technology. This article proposes a new network architecture based on a unicast/multicast protocol for data interaction between the SDN and IP network. There are three main contributions of this article. The first is that the architecture proposed in this article aims to reduce end-to-end transmission latency and packet loss. The second is to improve the flexibility of system configuration and precise control when the SND controller state changes. The third is to verify the feasibility of deploying network architecture in a real data center environment.

Energy-focused simulation of edge computing architectures in 5G networks

While cloud computing is crucial in processing data from devices with low computational power, the latency introduced by the Internet backhaul limits real-time applications. By situating computing resources at the network’s edge, edge computing offers low-latency services by offloading computations from high-performance computing (HPC) data centers to the edge servers, reducing wide Area network (WAN) strain. As a result, edge computing has unlocked opportunities for innovative applications that were previously unfeasible, such as connected vehicles or medical robotics. Nonetheless, deploying the infrastructure required to support edge computing services raises sustainability and energy consumption concerns. Consequently, the development of tools enabling researchers to explore innovative approaches to reducing the energy impact of edge computing is crucial. In this work, we present MintEDGE, a network simulator focused on the energy consumption of edge computing. Our simulator allows testing energy-saving approaches and task placement algorithms in realistic large-scale scenarios encompassing entire regions.

Application of high-voltage DC power supply in data center computer room

Application of high-voltage DC power supply in data center computer room

Dynamic Power Provisioning System for Fog Computing in IoT Environments

Large amounts of data are created from sensors in Internet of Things (IoT) services and applications. These data create a challenge in directing these data to the cloud, which needs extreme network bandwidth. Fog computing appears as a modern solution to overcome these challenges, where it can expand the cloud computing model to the boundary of the network, consequently adding a new class of services and applications with high-speed responses compared to the cloud. Cloud and fog computing propose huge amounts of resources for their clients and devices, especially in IoT environments. However, inactive resources and large number of applications and servers in cloud and fog computing data centers waste a huge amount of electricity. This paper will propose a Dynamic Power Provisioning (DPP) system in fog data centers, which consists of a multi-agent system that manages the power consumption for the fog resources in local data centers. The suggested DPP system will be tested by using the CloudSim and iFogsim tools. The outputs show that employing the DPP system in local fog data centers reduced the power consumption for fog resource providers.

Achieving resilient cities using data-driven energy transition: A statistical examination of energy policy effectiveness and community engagement

In an era where energy usage is increasingly critical due to the growing prevalence of cloud computing data centers, this study delves into the realm of dynamic energy management (DEM) within cloud data centers (CDCs) to foster efficient power utilization. Our research introduces a Dynamic System Management (DSM) framework tailored for CDCs, encompassing a dynamic voltage scaling (DVS) management unit, a load balancing unit, and a task scheduling unit (TSU). The TSU employs a stochastic Petri net-based planning procedure and advances a 'role-based' resource placement approach, optimizing task scheduling for enhanced system performance and energy efficiency. Through comprehensive simulation tests, we validate the proposed system scheme. Furthermore, this paper extends its focus to augment the resilience of energy policies in the context of smart cities and foster community engagement. Building upon the existing Biogeography-Based Optimization (BBO), we propose enhancements to fortify its robustness and capabilities. Notably, our modified BBO approach demonstrates remarkable reductions in power consumption—27%, 32%, and 39% less when compared to BBO, particle swarm optimization, and genetic algorithms, respectively. These findings underscore the pivotal role of data-driven energy transition and community engagement in building resilient cities for the future.

An algorithm for fast multiplication of elements in 2-groups based on the Zhegalkin polynomials

Network design for a multiprocessor computing system or data center is an important problem where the search for graph models that have attractive topological properties and allow the use of efficient routing algorithms is carried out. Cayley graphs have the indicated properties, in particular such as high symmetry, hierarchical structure, recursive design, high connectivity and fault tolerance. The definition of the Cayley graph implies that the vertices of the graph are elements of some algebraic group. Selecting a group and its generating elements allows us to obtain a graph that meets the necessary requirements for diameter, degree of vertices, number of nodes, etc. A large number of scientific articles and monographs are devoted to solving this problem. The goal of this work is to create an algorithm for fast multiplication of elements in finite 2-groups whose exponent is 2n. The first section of the article provides a theoretical justification for the algorithm for fast multiplication in finite 2-groups. It is shown that elements of these groups can be represented in the form of bit strings, and their multiplication is carried out based on the Zhegalkin polynomials. The second section presents the pseudocode of the algorithm on the basis of which the Zhegalkin polynomials are calculated. The third section demonstrates an example of obtaining the Zhegalkin polynomials for a two-generated group of exponent 4. In conclusion, the prospects for using the algorithm on the real hardware are discussed.

Performance and parameter optimization design of microchannel heat sink with different cavity and rib combinations

Microchannel heat sinks play a crucial role in dissipating heat in microelectronic systems in computer data centers. To enhance their thermal performance, this study proposes a combined microchannel design consisting of various cavity shapes and straight ribs, and analyzes its heat transfer and flow performance through numerical simulation. The heat transfer characteristics of microchannel heat sink with different straight rib structures are compared. Moreover, the four parameters including the relative length (α), relative width of the cavity (β), relative length of straight ribs (γ) and relative width of straight ribs (λ) are investigated, and the effects of Reynolds number variation on heat sink Nusselt number (Nu), friction coefficient (f) and thermal enhancement efficiency (η) are studied. The optimization process employs an artificial neural network and a multi-objective genetic algorithm to determine the optimal compromise solution and heat sink model, utilizing the Nusselt number and friction coefficient as evaluation indices. The results show that rectangular rounded straight rib is the straight rib structures with the best comprehensive thermal performance, with an average η approximately 9.7 % higher than that of the non-straight ribbed heat sink model. Furthermore, the variation in the four studied parameters yields distinct effects on the Nusselt number, friction coefficient, and thermal enhancement efficiency. Notably, the optimal performance is achieved when Nu = 13.59679 and f = 0.11855, with corresponding parameter values of α = 0.1575, β = 0.3931, γ = 0.0714, and λ = 1.2149. Ultimately, these results provide valuable insights into the structural optimization of microchannel heat sink cavity and straight rib combination models.

Mobile Edge Computing (MEC)-Enabled UAV Placement and Computation Efficiency Maximization in Disaster Scenario

Unmanned aerial vehicles (UAVs) are considered for disaster management to provide better coverage or help first responders provide advanced services. Furthermore, UAVs combined with mobile edge computing (MEC) capabilities, have recently emerged as a support for cloud computing by deploying UAV-assisted MEC at the network edge to reduce the latency and load of cloud computing data centers. We investigate the problem of maximizing the utility function by jointly optimizing the computing power, user associations, performance, duration, and location of user equipments (UEs) in UAV-assisted MEC networks in a disaster scenario. We considered several constraints including delay, quality of service, and coverage. The optimization problem is a mixed-integer non-linear programming problem. We propose a multi-stage offloading algorithm based on a learning algorithm and an interior-point method to obtain a viable solution. The simulation results obtained demonstrate the effectiveness of our proposed algorithm compared to existing schemes.

Treehouse: A Case For Carbon-Aware Datacenter Software

The end of Dennard scaling and the slowing of Moore's Law has put the energy use of datacenters on an unsustainable path. Datacenters are already a significant fraction of worldwide electricity use, with application demand scaling at a rapid rate. We argue that substantial reductions in the carbon intensity of datacenter computing are possible with a software-centric approach: by making energy and carbon visible to application developers on a fine-grained basis, by modifying system APIs to make it possible to make informed trade offs between performance and carbon emissions, and by raising the level of application programming to allow for flexible use of more energy efficient means of compute and storage. We also lay out a research agenda for systems software to reduce the carbon footprint of datacenter computing.

Investigating optimal 2D hydrodynamic modeling of a recent flash flood in a steep Norwegian river using high-performance computing

Abstract Efficient flood risk assessment and communication are essential for responding to increasingly recurrent flash floods. However, access to high-end data center computing is limited for stakeholders. This study evaluates the accuracy-speed trade-off of a hydraulic model by (i) assessing the potential acceleration of high-performance computing in PCs versus server-CPUs and GPUs, (ii) examining computing time evaluation and prediction indicators, and (iii) identifying variables controlling the computing time and their impact on the 2D hydrodynamic models' accuracy using an actual flash flood event as a benchmark. GPU-computing is found to be 130× and 55× faster than standard and parallelized CPU-computing, respectively, saving up to 99.5% of the computing time. The model's number of elements had the most significant impact, with <150,000 cells showing the best accuracy-speed trade-off. Using a PC equipped with a GPU enables almost real-time hydrodynamic information, democratizing flood data and facilitating interactive flood risk analysis.