Server Power Research Articles

The impact of distributed systems on the architecture and design of computer systems: advantages and challenges

A distributed system can encompass a variety of configurations, including mainframes, personal computers, workstations, and minicomputers. The varying degrees of software flexibility and the ability to execute tasks in parallel facilitate simultaneous data processing across multiple processors. The higher the resilience of an application, the quicker it can recover after a system failure. Organisations increasingly adopt distributed computing systems as they face increased data generation and demand for enhanced application performance. These systems enable businesses to scale effectively in response to growing data volumes. Integrating additional hardware into a distributed system is generally simpler than upgrading a centralised system reliant on powerful servers. Distributed systems comprise numerous nodes that collaborate towards a common objective. This article aims to provide a comprehensive overview of distributed systems, their architectural frameworks, and essential components. This study examines how distributed systems influence the architecture and design of computer systems. The research methods consist of reviewing existing literature and analysing case studies on implementing distributed systems. Key findings indicate that the evolution of distributed systems is ongoing, driven by emerging technologies and the increasing demand for efficient, scalable, and secure solutions. Innovations such as edge computing, blockchain technology, 5G, and the integration of AI and machine learning are among the notable trends shaping the future landscape of distributed systems. Looking ahead, designers and architects need to stay informed about these advancements to create reliable and adaptable distributed systems that can address the dynamic needs of users and organisations

Experimental and optimization research of the rack thermal environment based on the dynamic server power

Dynamic fluctuations in server power consumption significantly impact the thermal environment within the rack and the efficiency of computer room air conditioners (CRACs), necessitating thorough analysis it for safe and effective thermal management. Drawing on field experiment, this study presents an overview of the range and characteristics of variations in rack power consumption within a data center, revealing that 74.3 % of racks display periodic changes in power consumption. The change cycle has a diurnal pattern, and the dynamic changes in rack power consumption depend on the number of servers, the amount of task processing, and the nature of the customers. Subsequently, we designed the rack thermal environment experimental platform to analyze the response characteristics of the rack thermal environment to server power consumption changes and the variation rules of the thermal environment with various server layouts. The results show that a shift in server power severely affects the rack outlet temperature and is accompanied by a specific delay phenomenon. The near heat source effect, thermal buoyancy, and top heat accumulation primarily affect and form the rack thermal environment. It is recommended to place the server in the middle of the rack and with a 1U or 2U gap or uniform arrangement to create a better thermal environment. Finally, we propose three methods for optimizing the rack's thermal environment. Compared to the use of blind plates, the installation of air inlet deflectors, rear baffles, and top outlets can collectively reduce the hot spot temperature by up to 1.3 °C, 2.1 °C, and 0.4 °C, with the supply heat index (SHI) optimized at 0.184, 0.142, and 0.13. The return temperature index (RTI) was optimized at 88.99 %, 88.76 %, and 92.5 %, respectively.

A fuel cell powered rack prototype for data center application with waste heat recovery: design consideration and experimental verification

The combination of data center and PEMFC has been a promising data center zero-carbon emission solution. However, the relevant experimental studies are very rare and most work only solved the simple electrical coupling. This paper proposed a new in-rack data center servers and PEMFC coupling design scheme considering both the electrical coupling and thermal management coupling of the servers and PEMFC, and built a cabinet prototype for the first time. The complete multi-condition continuous experiment on the PEMFC system fully demonstrates the servers power supply stability of the prototype with the fast tested power response capability of 4.9 kW in 0.1 s. The experiments under rated condition (when the servers power is 5 kW) were conducted and the carbon emission level and efficiency performance of the prototype are analyzed. Under the rated condition, the highest prototype CHP efficiency is 64.1 % while the prototype can provide 41.6 ℃ hot water with 3.04 L·min-1 and the lowest prototype CHP efficiency is 59.8 % while the prototype can provide 61.1 ℃ hot water with 1.33 L·min-1. The prototype maintains the zero-carbon (or negative carbon) emission of servers. The average operation carbon emission rate is -35 g·h-1. Compared with the full grid supporting data center, the power average prototype operation carbon reduction rate is 0.13 g·s-1·kW-1. The application potential of the proposed data center and PEMFC coupling scheme is demonstrated with the prototype experiment results from the advantage of carbon emission and heat recovery potential.

Integrated analysis of reliability, power, and performance for IoT devices and servers

The Internet of Things (IoT) is currently widely used in various sectors and spaces. IoT devices are becoming small yet powerful servers and perform server-like functions. Reliability is a critical aspect in both IoT devices and servers, as they work together to create a robust and dependable IoT ecosystem. Power and performance are two other major considerations of an IoT system. Modeling, analysis, evaluation, and optimization of reliability, power, and performance for IoT devices and servers are major components in IoT systems development and deployment. In this paper, we conduct an integrated study of reliability, power, and performance for IoT devices and servers by mathematically rigorous modeling and analysis. The contributions of the paper can be summarized as follows. We establish a continuous-time Markov chain (CTMC) model that incorporates server failure rate, server repair rate, task arrival rate, and task processing rate. Using such an analytical model, we can calculate the server availability, the average task response time, and the average power consumption. We point out that there is an optimal server speed that minimizes the power-time product and a combined cost-performance metric of power, performance, and reliability. We show the impact of server reliability on response time, power consumption, server utilization, and the power-performance tradeoff. To the best of the author’s knowledge, this is the first paper that takes a combined approach to modeling and analysis of reliability, power, and performance for IoT devices and servers. It has been noticed that there has been little such theoretically solid investigation in the existing literature. Therefore, this paper has made tangible contributions and significant advances in the joint understanding of reliability, power, performance, and their interplay in IoT devices and servers quantitatively and mathematically.

UAV‐aided distribution line inspection using double‐layer offloading mechanism

AbstractWith the continuous growth of electricity demand, the safe and stable operation of distribution lines is crucial for power transportation. Unmanned aerial vehicle (UAV) inspection has been widely used for the maintenance and repair of distribution lines. Due to the limitations of computational power and endurance, it is difficult for UAVs to independently complete data processing. Combined with mobile edge computing (MEC), this paper proposes a computing offloading strategy based on multi‐agent reinforcement learning and double‐layer offloading mechanism, which can further utilize the computing power of non‐task devices and edge servers. Firstly, three‐layer system architecture, named MEC‐U‐NTDC (MEC‐UAV‐Non‐task Device Cloud), is built. Secondly, double‐layer offloading mechanism is designed to comprehensively utilize the computing power of edge servers and neighbouring non‐task devices. Finally, a multi‐agent algorithm DLMQMIX is proposed to minimize the total cost for UAV inspection. Simulation experiments show that the proposed algorithm can effectively solve the task offloading problem of UAV‐aided distribution line inspection, and compared with algorithms such as PSO, GA, and QMIX, it performs better in terms of average delay, system cost, and load balancing, achieving a smaller total system cost.

A novel virtual machine consolidation algorithm with server power mode management for energy-efficient cloud data centers

A novel virtual machine consolidation algorithm with server power mode management for energy-efficient cloud data centers

Employing machine learning (ML) and explainable artificial intelligence (XAI) to predict and explain suicidal ideation among patients with prostate cancer.

e23086 Background: It is anticipated that implementing machine learning (ML) strategies will impact the clinical decision-making process in the future. This study aimed to compare the accuracy of various machine learning algorithms in predicting suicidal ideation in patients who were receiving treatment for prostate cancer (PC). Further, to predict suicidal ideation and explain the burden of illness (BOI) by utilizing ML algorithms and XAI-based interpretability. Methods: We analyzed the United States 2017 National Inpatient Sample database for patients hospitalized for PC using linear and non-linear ML methods. Using techniques such as forward selection (FS) and backward elimination (BE), Random Forest (RF), decision trees, Multivariate Adaptive Regression Splines, and Gradient Boosting Machine (GBM), we determined subsets and features. We used linear and non-linear MLs-- Lasso, Ridge, RF, and Neural Networks (NN). A 70% and 30% partitioning was performed on the training and test datasets. The performance of the model was tested using discrimination (C-statistics), the Receiver-Operating Characteristics (ROC) curve, and the Hosmer-Lemeshow tests on both the training data and the test data. Several XAI approaches, including permutation significance, global surrogate marker, feature interpretation, interactivity, and local interpretability, were also utilized to examine the BOI (Length of stay (LOS)) among PC cancer cohort with suicide ideation. Results: We identified 680 patients with suicidal ideation among 208,730 PC patients. The most important variable derived through FS and BE were depression, drug abuse, psychiatric disorder, alcohol dependence, race, Medicaid beneficiaries, cardiac arrhythmias, metastasis, weight loss, congestive heart failure, and un-complicated hypertension, further confirmed through the GBM and other methods. There was successful documentation of demographic, socioeconomic, and clinical characteristics through feature selection approaches. Linear models, Lasso, and Ridge demonstrated an excellent area under the ROC for all cohorts ( > 0.9 in train and test datasets). On the other hand, tree-based models, such as RF, performed poorly (AUC 0.72 for train and 0.66 for test). The findings of the XAI methodologies showed that among PC suicide ideation, higher BOI was associated with patients staying in low-income neighborhoods, Whites, Medicaid beneficiaries, and those with weight loss. Conclusions: We showcase the increasing capability of predictive analytics, particularly XAI, in predicting suicidal ideation. The combination of the vast amount of data generated by the digitalization of healthcare systems with the advanced processing power of modern servers enables the development of machine learning-based care pathways, which has the potential to revolutionize clinical decision-making in cancer.

OptimML: Joint Control of Inference Latency and Server Power Consumption for ML Performance Optimization

Power capping is an important technique for high-density servers to safely oversubscribe the power infrastructure in a data center. However, power capping is commonly accomplished by dynamically lowering the server processors’ frequency levels, which can result in degraded application performance. For servers that run important machine learning (ML) applications with Service-Level Objective (SLO) requirements, inference performance such as recognition accuracy must be optimized within a certain latency constraint, which demands high server performance. In order to achieve the best inference accuracy under the desired latency and server power constraints, this paper proposes OptimML, a multi-input-multi-output (MIMO) control framework that jointly controls both inference latency and server power consumption, by flexibly adjusting the machine learning model size (and so its required computing resources) when server frequency needs to be lowered for power capping. Our results on a hardware testbed with widely adopted ML framework (including PyTorch, TensorFlow, and MXNet) show that OptimML achieves higher inference accuracy compared with several well-designed baselines, while respecting both latency and power constraints. Furthermore, an adaptive control scheme with online model switching and estimation is designed to achieve analytic assurance of control accuracy and system stability, even in the face of significant workload/hardware variations.

A Differential Evolution Offloading Strategy for Latency and Privacy Sensitive Tasks with Federated Local-edge-cloud Collaboration

Due to an explosive growth in mobile devices and the rapid evolution of wireless communication technologies, local-edge-cloud computing is becoming an attractive solution for providing a higher-quality service by exploiting the multi-computation power of mobile devices, edge servers and cloud. However, as the tasks are latency and privacy sensitive, highly credible task offloading becomes a crucial problem in a local-edge-cloud orchestrated computing system. In this paper, we study the computation offloading problem for latency and privacy sensitive tasks in a hierarchical local-edge-cloud network by using federated learning method. Our goal is to minimize the operational time of latency-sensitive tasks requested by mobile devices that have data privacy concerns, while each task can be executed under local, edge or cloud computing mode with no need to rely on privacy data. We first build system models to analyze the latency incurred under different computing modes, and then develop a constrained optimization problem to minimize the latency consumed by the federated offloading collaboration. A Hierarchical Federated Averaging method based on Differential Evolution algorithm (HierFAVG-DE) is proposed for solving the problem in-hand, and extensive simulations are conducted to verify the superiority of our approach.

On Efficient Parallel Secure Outsourcing of Modular Exponentiation to Cloud for IoT Applications

Modular exponentiation is crucial for secure data exchange in cryptography, especially for resource-constrained Internet of Things (IoT) devices. These devices often rely on third-party servers to handle computationally intensive tasks like modular exponentiation. However, existing outsourcing solutions for the RSA algorithm may have security vulnerabilities. This work identifies a critical flaw in a recent outsourcing protocol for RSA proposed by Hu et al. We demonstrate how this flaw compromises the security of the entire RSA system. Subsequently, we propose a robust solution that strengthens the RSA algorithm and mitigates the identified vulnerability. Furthermore, our solution remains resilient against existing lattice-based attacks. The proposed fix offers a more secure and efficient way for IoT devices to leverage the power of third-party servers while maintaining data integrity and confidentiality. An extensive performance evaluation confirms that our solution offers comparable efficiency while significantly enhancing security compared to existing approaches.

FLSTRA: Federated Learning in Stratosphere

We propose a federated learning (FL) in stratosphere (FLSTRA) system,where a high altitude platformstation (HAPS) facilitates a large number of terrestrial clients to collaboratively learn a global model without sharing the training data. FLSTRA overcomes the challenges faced by FL in terrestrial networks, such as slow convergence and high communication delay due to limited client participation and multi-hop communications. HAPS leverages its altitude and size to allow the participation of more clients with line-of-sight (LOS) links and the placement of a powerful server. However, handling many clients at once introduces computing and transmission delays. Thus, we aim to obtain a delay-accuracy trade-off for FLSTRA. Specifically, we first develop a joint client selection and resource allocation algorithm for uplink and downlink to minimize the FL delay subject to the energy and quality-of-service (QoS) constraints. Second, we propose a communication and computation resource-aware (CCRA-FL) algorithmto achieve the target FL accuracy while deriving an upper bound for its convergence rate. The formulated problemis non-convex; thus, we propose an iterative algorithmto solve it. Simulation results demonstrate the effectiveness of the proposed FLSTRA system, compared to terrestrial benchmarks, in terms of FL delay and accuracy.

SecBERT: Privacy-preserving pre-training based neural network inference system

Pre-trained models such as BERT have made great achievements in natural language processing tasks in recent years. In this paper, we investigate the privacy-preserving pre-training based neural network inference in a two-server framework based on additive secret sharing technique. Our protocol allows a resource-restrained client to request two powerful servers to cooperatively process the natural processing tasks without revealing any useful information about its data. We first design a series of secure sub-protocols for non-linear functions used in BERT model. These sub-protocols are expected to have broad applications and of independent interest. Based on the building sub-protocols, we propose SecBERT, a privacy-preserving pre-training based neural network inference protocol. SecBERT is the first cryptographically secure privacy-preserving pre-training based neural network inference protocol. We show security, efficiency and accuracy of SecBERT protocol through comprehensive theoretical analysis and experiments.

AMIS-MU: Edge Computing Based Adaptive Video Streaming for Multiple Mobile Users

The increasing demand for online high-quality video streaming has brought huge challenges to the traditional client-server video streaming systems due to the high feedback delay, rigorous bandwidth requirement, and the lack of a mechanism of centralized resource management between users. In this work, we propose AMIS-MU, an edge computing-based mobile video streaming system that optimizes the watching experience of users via playback adaptation and channel resource allocation. AMIS-MU fully explores the power of edge servers from three perspectives. First, by pre-caching videos from the cloud, AMIS-MU analyzes video contents at the edge, and achieves a nearly imperceptible content-based playback speed adaptation. Second, as the edge server controls the channel resources of users in a centralized fashion, AMIS-MU adaptively updates the channel configuration to optimize the overall watching experience. Last, the plenty of computational power available at the edge enables a more intelligent playback control by using deep reinforcement learning (DRL). We propose a novel usage of DRL which significantly reduces the complexity of the cross-layer joint optimization problem and solve the non-convex channel resource allocation problem by Lyapunov optimization. Experiments show that AMIS-MU outperforms other existing algorithms in terms of average QoE and fairness.

A Hybrid Current Reference Control Method for PFC Converter in Server Power Supply

A Hybrid Current Reference Control Method for PFC Converter in Server Power Supply

A Truthful Reverse Auction Mechanism for Federated Learning Utility Maximization Resource Allocation in Edge–Cloud Collaboration

Federated learning is a promising technique in cloud computing and edge computing environments, and designing a reasonable resource allocation scheme for federated learning is particularly important. In this paper, we propose an auction mechanism for federated learning resource allocation in the edge–cloud collaborative environment, which can motivate data owners to participate in federated learning and effectively utilize the resources and computing power of edge servers, thereby reducing the pressure on cloud services. Specifically, we formulate the federated learning platform data value maximization problem as an integer programming model with multiple constraints, develop a resource allocation algorithm based on the monotone submodular value function, devise a payment algorithm based on critical price theory and demonstrate that the mechanism satisfies truthfulness and individual rationality.

Comprehensive Analysis on Power Factor Correction Circuit

With the development of industrial power electronics, the current era can be called the "world energy era". Energy has never been more important, especially with the growing demand for today's "green" energy sources. power factor correction circuits have become an integral part of most current AC voltage input power applications, whether it is automotive battery power, portable energy storage (such as power banks), traditional server power, or lighting systems. However, the basic principles of power factor correction converters are rarely described in detail. This paper will summarize and illustrate the basic principle of the power factor correction circuit, and first introduce the research background of power factor correction technology; At the same time, three operating modes of power factor correction circuit are analyzed: continuous on-mode (CCM), critical on-mode (CRM) and discontinuous on-mode (DCM), and their advantages and disadvantages are to be further discussed. In addition, the PFC converters with optimized topology and excellent performance will be introduced, and the application prospect and prospect of three-phase PFC converters will be presented.

Frankenstein: fast and lightweight call graph generation for software builds

Call Graphs are a rich data source and form the foundation for advanced static analyses that can, for example, detect security vulnerabilities or dead code. This information is invaluable when it is immediately available, such as in the output of a build system. Call Graph generation is a whole-program analysis: not just the application, but also all its dependencies are processed together. Recent work has shown that even advanced static analyses can use summarization techniques to substantially improve runtime; however, existing analyses focus on soundness, and as such remain very expensive. When executed in the build system, which typically has limited resources, even powerful servers suffer from slow build times, rendering these analyses impractical in today’s fast-paced development. In this paper, we aim to strike a balance between improving static analyses while remaining practical for use cases that require quick results in low-resource environments. We propose a summarization-based implementation of a Class-Hierarchy Analysis algorithm for call graph generation of Java programs. Our approach leverages the fact that dependency sets often do not change between builds: we can generate call graphs for these dependencies, cache their generation for subsequent builds, and using a novel stitching algorithm, Frankenstein, merge all partial results into a complete call graph for the whole program. Our evaluation results show that this lightweight approach can substantially outperform existing frameworks. In terms of speed improvements, Frankenstein surpasses the baselines by up to 38%, requiring an average of just 388 Megabytes of memory. This makes the proposed approach practical for build systems with limited memory resources. Despite these optimizations, our generated call graphs maintain a near-identical set of edges when compared to the baselines, achieving an F1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$_{1}$$\\end{document} score of up to 0.98. This summarization-based approach for call graph generation paves the way for using extended static analyses in build processes.

Container Power Consumption Prediction Based on GBRT-PL for Edge Servers in Smart City

Edge computing and IoT devices have been widely deployed in smart city applications due to the rapid promotion and implementation of 5G communication technology. Limited by power supply and hardware computing capability, applications on edge servers are mostly deployed and run in the form of container micro-services to improve resource utilization. Therefore, identifying the power consumption at the container granularity is of great importance for the load scheduling and service quality assurance of edge servers. In this paper, a GBRT-PL-based container power prediction method is proposed. Performance metrics with a strong correlation between container and server power consumption are selected for power consumption modeling. To effectively suit the nonlinear relationship between container performance metrics and server power consumption, the integrated predictive capabilities of multiple regression trees and the segmented linear model of single regression tree leaf nodes are applied. The data from the experiments prove that the GBRT-PL model predicts power consumption more accurately than other models for single and multiple container groups. The highest average relative error rate in the four multi-container group tests is 6.72%, whereas the highest relative error rate in the 90% quantile is 11.66%. In addition, it can accurately predict the majority of power consumption peaks, which contributes to the precise detection of power consumption anomalies.

Optimization of Personal Credit Evaluation Based on a Federated Deep Learning Model

Financial institutions utilize data for the intelligent assessment of personal credit. However, the privacy of financial data is gradually increasing, and the training data of a single financial institution may exhibit problems regarding low data volume and poor data quality. Herein, by fusing federated learning with deep learning (FL-DL), we innovatively propose a dynamic communication algorithm and an adaptive aggregation algorithm as means of effectively solving the following problems, which are associated with personal credit evaluation: data privacy protection, distributed computing, and distributed storage. The dynamic communication algorithm utilizes a combination of fixed communication intervals and constrained variable intervals, which enables the federated system to utilize multiple communication intervals in a single learning task; thus, the performance of personal credit assessment models is enhanced. The adaptive aggregation algorithm proposes a novel aggregation weight formula. This algorithm enables the aggregation weights to be automatically updated, and it enhances the accuracy of individual credit assessment by exploiting the interplay between global and local models, which entails placing an additional but small computational burden on the powerful server side rather than on the resource-constrained client side. Finally, with regard to both algorithms and the FL-DL model, experiments and analyses are conducted using Lending Club financial company data; the results of the analysis indicate that both algorithms outperform the algorithms that are being compared and that the FL-DL model outperforms the advanced learning model.

Blockchain Based Decentralized Web Hosting System

Abstract: In today’s digital era, the internet heavily depends on a centralized framework where powerful servers support content on well-known platforms such as Facebook and Twitter. Although this centralization is proficient in content delivery, it comes with significant drawbacks, including a single point of failure and vulnerability to censorship. The need for a decentralized web hosting structure is growing due to its promise of content accessibility regardless of a user’s geographical location, while still maintaining the speed and reliability of conventional web services. Currently, the internet’s structure is defined by centralization, introducing risks related to cyber-attacks and service disturbances. For example, even a brief disruption to a large platform like Facebook can leave millions without internet access. Additionally, large corporations and government bodies possess significant control over the internet, enabling them to alter and censor content. To confront these problems, it becomes essential to develop a decentralized web structure that is publicly owned. This study delves into the issues arising from the centralized internet and investigates potential solutions by utilizing blockchain technology to decentralize web structure, subsequently improving traditional web services and fortifying the internet’s robustness and accessibility