Surviving failures in bandwidth-constrained datacenters

Failures are normal rather than exceptional in cloud computing environments, high fault tolerance issue is one of the major obstacles for opening up a new era of high serviceability cloud computing as fault tolerance plays a key role in ensuring cloud serviceability. Fault tolerant service is an essential part of Service Level Objectives (SLOs) in clouds. To achieve high level of cloud serviceability and to meet high level of cloud SLOs, a foolproof fault tolerance strategy is needed. In this paper, the definitions of fault, error, and failure in a cloud are given, and the principles for high fault tolerance objectives are systematically analyzed by referring to the fault tolerance theories suitable for large-scale distributed computing environments. Based on the principles and semantics of cloud fault tolerance, a dynamic adaptive fault tolerance strategy DAFT is put forward. It includes: (i) analyzing the mathematical relationship between different failure rates and two different fault tolerance strategies, which are checkpointing fault tolerance strategy and data replication fault tolerance strategy; (ii) building a dynamic adaptive checkpointing fault tolerance model and a dynamic adaptive replication fault tolerance model by combining the two fault tolerance models together to maximize the serviceability and meet the SLOs; and (iii) evaluating the dynamic adaptive fault tolerance strategy under various conditions in large-scale cloud data centers and consider different system centric parameters, such as fault tolerance degree, fault tolerance overhead, response time, etc. Theoretical as well as experimental results conclusively demonstrate that the dynamic adaptive fault tolerance strategy DAFT has high potential as it provides efficient fault tolerance enhancements, significant cloud serviceability improvement, and great SLOs satisfaction. It efficiently and effectively achieves a trade-off for fault tolerance objectives in cloud computing environments.

Most previous work on fault-tolerant (FT) multiprocessor design has concentrated on deterministic k-fault-tolerant (k-FT) designs in which exactly k spare processors and some spare switches and links are added to construct multiprocessors that can tolerate any k processor faults. However, after k faults are reconfigured around, much of the extra links and switches can remain unutilized. We show how to use the node-covering principle of Dutt and Hayes (1992) and error correcting codes in order to construct probabilistic designs with very high average fault tolerance but low wiring and switch overhead. This design methodology is applicable to any multiprocessor interconnection topology. We also obtain the deterministic fault tolerance for these designs and develop efficient layout strategies for them. >

Fault tolerance, performance, and throughput have been major areas of research and development since the evolution of large-scale networks. Internet-based applications are rapidly growing, including large-scale computations, search engines, high-definition video streaming, e-commerce, and video on demand. In recent years, energy efficiency and fault tolerance have gained significant importance in data center networks and various studies directed the attention towards green computing. Data centers consume a huge amount of energy and various architectures and techniques have been proposed to improve the energy efficiency of data centers. However, there is a tradeoff between energy efficiency and fault tolerance. The objective of this study is to highlight a better tradeoff between the two extremes: (a) high energy efficiency and (b) ensuring high availability through fault tolerance and redundancy. The main objective of the proposed Energy-Aware Fault-Tolerant (EAFT) approach is to keep one level of redundancy for fault tolerance while scheduling resources for energy efficiency. The resultant energy-efficient data center network provides availability as well as fault tolerance at reduced operating cost. The main contributions of this article are: (a) we propose an Energy-Aware Fault-Tolerant (EAFT) data center network scheduler; (b) we compare EAFT with energy efficient resource scheduling techniques to provide analysis of parameters such as, workload distribution, average task per servers, and energy consumption; and (c) we highlight effects of energy efficiency techniques on the network performance of the data center.

Nowadays, there exist many structured P2P network models, but most of them cannot provide explicit semantic locality and topology consistency, and they share the shortcoming of low fault tolerance. In this paper, Rings, a novel model is presented. It can provide high routing efficiency with logarithmic order and has strong fault tolerance: in dynamic network environment, especially when network partitioning takes place, it could ensure functionality of most nodes. Besides, Rings could support explicit semantic locality in taxonomy hierarchical semantic environment. On the other hand, Rings also can be applied in normal network environment. Therefore it's a generic model. What's more, it has the topological awareness. With all the above features, Rings becomes an outstanding structured P2P network model. Extensive experiments are conducted to evaluate the performance and verify the advantages of Rings. The results show that Rings has good performance.

This work targets design of a robust universal logic gate in Quantum-dot cellular automata (RUQCA) that realizes majority and minority functions simultaneously with high fault tolerance. An alternative tile structure of QCA with hybrid cell orientations is formulated to maximize its throughput. The characterization of defective behaviour of RUQCA gate under cell deposition, cell misplacement and cell misalignment defects is investigated. The results show that the proposed RUQCA gate has very high fault coverage of 95%. The synthesis of fault tolerant multiplexer using RUQCA, with 90% fault tolerance, establishes the effectiveness of RUQCA.

This paper presents a new fault-tolerant magnetic-geared permanent-magnet (PM) motor for direct-drive applications. The key of the proposed motor is to integrate a magnetic gear into a fault-tolerant PM motor, in which the pole-pairs of PMs on both sides of the inner rotor are unequal. Hence, it can incorporate the merits of high torque-to-volume ratio, high gearing ratio, and low-speed high-torque operation with steady speed and high fault tolerance. A new design idea is reported, and electromagnetic performance of the proposed motor is analyzed by using finite element method. Finally, the prototyped motor verifies with theoretical analysis.

The new power system has the characteristics of “double height”, which mainly includes the introduction of wind power/photovoltaic and other power electronic devices on the basis of conventional power equipment, and the addition of typical converter valve hall internal converter transformer/valve tower/wall bushing/converter Bushing/super large voltage equalizing ball and other power equipment. For example, a large number of the power equipment need real-time operation and maintenance, it is urgent to develop efficient/practical/convenient intelligent operation and inspection technology for power equipment. Based on this, this paper combines image processing with deep learning technology, in which the image data includes high-definition digital photos/UV imager/infrared imager photos, and the deep learning algorithm mainly includes neural network and wavelet function processing. Through the above artificial intelligence algorithm processing, automatic extraction function of power equipment in image database and the comparative analysis function of historical image database are realized. Based on the above joint algorithm, it can effectively reduce the number of out detection of power equipment inspectors, automatically compare and analyze the massive image data acquisition and the historical image database, effectively discover the gradually strengthening points of UV discharge, the gradually increasing points of infrared imaging temperature or the gradually distorted points of displacement morphology, call high-definition digital photos based on the above information to mine the abnormal position of the state, and enlarge it to obtain the defect type. This paper attempts to provide the technical means for the intelligent operation inspection of the power equipment from the perspective of image processing. This technology has the broad application space and good theoretical research and engineering application value.

During the operation and maintenance of power equipment, a large amount of text data is accumulated, and it is of great importance to mine valuable information and evaluate the operation status of the equipment. Among them, named entity recognition technology is a key prerequisite for downstream tasks. However, with the development of natural language processing technology, while improving the accuracy of entity recognition, the existing models are gradually unable to meet the requirements of time and equipment cost for model training in practice. In this paper, we propose a low-cost ALBERT-BiLSTM-CRF-based named entity recognition model applicable to power equipment defective text. The model achieves an F1 score of 92.47% in entity recognition in the power domain, outperforming the benchmark BERT model performance in terms of time cost and effect.

The information and communication technology nowadays more than ever depends on the Internet and cloud computing, so that the data centers (DCs) have been converted to a constitutive unit of the cloud computing. A DC is composed of two primary parts: servers and Data Center Networks (DCNs). Robustness and scalability are two major challenges of the DCNs that are expanded based on two strategies, scale-out, and scale-up. This paper is distinctive from the related studies in two aspects. The first one is to simultaneously focus on both the scalability and the robustness challenges of the DCNs. For this purpose, we will concentrate on the comparison of robustness in the scalable models of these networks. The second one is, despite the previous work that only evaluated the DCN robustness under topological changes, we evaluated the robustness and fault tolerance against three types of unexpected changes in topology, traffic, and COI (community of interest) in the present work. Hence, we have chosen the network criticality (NC) as a graph-theoretic metric for analyzing DCN robustness. Afterward, we compare some structural and spectral graph metrics with NC among some well-known DCNs, and their scale-out and scale-up. Our results are useful to select the appropriate scaling strategy with the goal of maximizing the robustness of existing DCNs and provide a guideline for designing the new robust and scalable DCN.

The author reviews number-theoretic techniques for achieving hardware modularity in order to facilitate high data rates, testability, reliability, and fault tolerance, in VLSI digital signal processing systems. The theory of RNS (residue number system) error detection and correction is reviewed, and the special properties of modular systems are discussed for providing a rich environment for fault tolerant designs. Questions of reliability and fault tolerance on both the integrated circuit level and higher systems levels are discussed. The design of a convolutional back-projection digital processor for synthetic aperture radar (SAR) image processing is used as an example to investigate appropriate interactions between circuit-level error checking and system-level fault tolerance. >

Triple redundant fault tolerance: A hardware-implemented approach

In recent years, data center network is used for transmission, storage and processing of big data, which plays an important role for applications in cloud computing and CDN distribution. Network topology and routing algorithm are its core research content and key technical issues. The traditional network topology is difficult to guarantee the quality of service in scalability and fault tolerance. The server-centric data center network topology can ensure the scale of the data center network by recursively increasing the number of network nodes and links. relative to the Dcell, BCube, and BCCC typical network topology, PTNet network as a typical representative of a new type of the server-centric data center network topology, which has more advantages in scalability, fault tolerance and so on. Multicast and broadcast in data center network have more application scenarios and use value. For example, the video conference online, multimedia remote education and other development are inseparable from the application and promotion of network multicast and broadcast. So it is necessary to research the routing algorithms of multicast and broadcast in the network. Based on the deep research of PTNet network, this paper analyzes and researches the network topology, multicast and broadcast routing algorithm.

The continuous deterioration of air and water quality worldwide poses severe threats to human health, biodiversity, and environmental sustainability. Rapid urbanization, industrial activities, and increasing vehicular emissions have escalated pollution levels, especially in developing nations like India. Traditional centralized environmental monitoring systems are often inefficient due to high data transmission delays, network congestion, and limited scalability, making them inadequate for real-time monitoring and response. The integration of Edge Computing and Internet of Things (IoT) technologies provides an innovative framework to overcome these challenges. Edge devices enable local data processing near IoT sensors, minimizing latency, reducing bandwidth usage, and ensuring timely and efficient monitoring outcomes. This paper reviews the integration of Edge Computing and IoT for real-time air and water quality monitoring systems, highlighting their architecture, implementation strategies, benefits, challenges, and future directions.We present recent real-time data from major Indian cities showing Delhi’s Air Quality Index (AQI) at 245, significantly exceeding safe levels, and Kolkata’s Water Quality Index (WQI) at 50, indicating poor water conditions. The paper includes system architecture diagrams, comparative data tables, and graphical representations to provide a comprehensive understanding of these systems. Key challenges such as energy efficiency, device interoperability, data security, and scalability are analyzed, and solutions like federated learning and blockchain integration are explored. Future research opportunities in predictive environmental analytics and citizen-centric monitoring are identified. This review aims to assist researchers, environmental policymakers, and smart city developers in designing resilient, scalable, and real-time environmental monitoring systems leveraging edge–IoT synergies.

“Emission Peak and Carbon Neutrality” has become one of the important goals of my country's future development. At this stage, there are three major challenges in achieving the dual-carbon goal in the power sector: difficult to fine-tune carbon emissions statistics, difficult to scientific quantitative analysis, difficult to individualize carbon emissions management, and lack of quantitative algorithms for the carbon footprint of power main equipment. This paper analyzes the research status of carbon emission accounting methods in the electric power field, and puts forward a quantitative calculation method for the carbon footprint of the main equipment of electric power. The statistical range of carbon foot-print of the main equipment of electric power system is defined, the carbon footprint quantification method is formulated, and the carbon footprint optimization suggestions are put forward. The carbon footprint statistical analysis of the whole life cycle of power equipment manufacturing, operation and maintenance, decommissioning and recycling has been realized. Based on this algorithm, a visualization platform for the carbon footprint of the main equipment of electric power is developed, and the carbon footprint of the main equipment of an UHV station is traced back to the station based on the main equipment, and the visualization of the carbon footprint of the main equipment of electric power is realized.

Multiphase motors have high reliability and good fault tolerance ability, and are widely used in electric drive systems such as electric vehicles, ships and aircrafts. In these systems, the peak power of the motor is high, but it works in the light load state for a long time. When the motor is in fault-tolerant state, the heating of windings and iron core is unbalanced, which decreases the motor performance. In this paper, an asymmetric winding multi-phase motor structure is proposed. The performance with different slot pole combination and winding distribution are compared. The torque, efficiency and temperature rise under different working modes such as normal operation, whole set of winding cut off and single-phase winding cut off are analyzed. The results show that asymmetric windings can improve the efficiency of the system under longterm light load conditions. The single-layer and double-layer winding motors have lower torque and higher ripple under fault-tolerant conditions; and the performance degradation can be suppressed by using multi-layer windings, and the suitable slot pole combination for multi-layer windings is obtained through comparative analysis.