检索结果-内蒙古大学图书馆

Lauca: A Workload Duplicator for Benchmarking Transactional Database Performance

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第7期36卷 3180-3194页

作者： Weng, Siyang Wang, Qingshuai Qu, Luyi Zhang, Rong Cai, Peng Qian, Weining Zhou, Aoying East China Normal Univ Sch Data Sci & Engn Shanghai 200050 Peoples R China

Generating synthetic workloads is essential and critical to the performance evaluation of database systems. When benchmarking database performance for a specific application, the similarity between synthetic workloads and real application workloads determines the credibility of the evaluation results. However, it meets a great challenge to catch workload characteristics for a target online transaction processing (OLTP) application considering the complexity of transaction executions. To address this problem, we propose a workload duplicator (Lauca) that can generate synthetic workloads with highly similar performance metrics compared to a specific application on both centralized and distributed databases. By carefully studying the application-driven workload generation problem, we present Transaction Logic, Data Access Distribution and Partition Access Distribution to characterize runtime workloads and propose novel generation algorithms to guarantee the high fidelity of synthetic workloads. To the best of our knowledge, Lauca is the first application-driven transactional workload generator. We conduct extensive experiments based on TPC-C, SmallBank and YCSB on both centralized and distributed databases. The experimental results show that Lauca consistently generates high-quality synthetic workloads.

关键词： databases Benchmark testing Runtime distributed databases Throughput Semantics Performance evaluation synthetic workload OLTP applications

来源：评论

学校读者我要写书评

暂无评论

Revisiting Trust Management in the Data Economy: A Road Map

引用

IEEE INTERNET COMPUTING 2024年第4期28卷 21-29页

作者： Ardagnaa, Claudio A. Bena, Nicola Bennani, Nadia Ghedira-Guegan, Chirine Grecchi, Nicolo Vargas-Solar, Genoveva Univ Milan I-20133 Milan Italy INSA Lyon LIRIS CNRS UMR5205 F-69621 Villeurbanne France Univ Lyon 3 iaelyon Sch Management INSA Lyon LIRISUMR5205 F-69008 Lyon France Univ Milan Syst & Network Secur I-20133 Milan Italy CNRS UMR5205 F-69622 Villeurbanne France

In the last two decades, multiple information and communications technology evolutions have boosted the ability to collect and analyze vast numbers of data (on the order of zettabytes). Collectively, they have paved the way for the so-called data economy, revolutionizing most sectors of our society, including health care, transportation, and grids. At the core of this revolution, distributed data-intensive applications compose services operated by multiple parties in the cloud-edge continuum;they process, manage, and exchange massive numbers of data at an unprecedented rate. However, data hold little value without adequate data protection. Traditional solutions, which aim to balance data quality and protection, are insufficient to address the peculiarities of the data economy, including trustworthy data sharing and management, composite service support, and multiparty data lifecycle. This article analyzes how trust management systems (TMSs) can regain the lead in supporting trustworthy data-intensive applications, discussing current challenges and proposing a road map for new-generation TMSs in the data economy.

关键词： Trust management distributed databases Computer architecture Europe Protocols Data protection Data models Information and communication technology Data integrity

来源：评论

学校读者我要写书评

暂无评论

distributed Blind Source Separation Based on FastICA

引用

IEEE SIGNAL PROCESSING LETTERS 2025年 32卷 881-885页

作者： Musluoglu, Cem Ates Bertrand, Alexander Katholieke Univ Leuven Stadius Ctr Dynam Syst Dept Elect Engn ESAT Signal Proc & Data Analyt B-3001 Leuven Belgium Leuven AI KU Leuven Inst AI B-3001 Leuven Belgium

With the emergence of wireless sensor networks (WSNs), many traditional signal processing tasks are required to be computed in a distributed fashion, without transmissions of the raw data to a centralized processing unit, due to the limited energy and bandwidth resources available to the sensors. In this paper, we propose a distributed independent component analysis (ICA) algorithm, which aims at identifying the original signal sources based on observations of their mixtures measured at various sensor nodes. One of the most commonly used ICA algorithms is known as FastICA, which requires a spatial pre-whitening operation in the first step of the algorithm. Such a pre-whitening across all nodes of a WSN is impossible in a bandwidth-constrained distributed setting as it requires to correlate each channel with each other channel in the WSN. We show that an explicit network-wide pre-whitening step can be circumvented by leveraging the properties of the so-called distributed Adaptive Signal Fusion (DASF) framework. Despite the lack of such a network-wide pre-whitening, we can still obtain the $Q$ least Gaussian independent components of the centralized ICA solution, where $Q$ scales linearly with the required communication load.

关键词： Signal processing algorithms Wireless sensor networks Sensors distributed databases Europe Artificial intelligence Sensor fusion Optimization Indexes Fitting distributed optimization distributed spatial filtering independent component analysis

来源：评论

学校读者我要写书评

暂无评论

Adaptive Block-Wise Regularization and Knowledge Distillation for Enhancing Federated Learning

引用

IEEE-ACM TRANSACTIONS ON NETWORKING 2024年第1期32卷 791-805页

作者： Liu, Jianchun Zeng, Qingmin Xu, Hongli Xu, Yang Wang, Zhiyuan Huang, He Univ Sci & Technol China Sch Comp Sci & Technol Hefei 230027 Anhui Peoples R China Univ Sci & Technol China Suzhou Inst Adv Res Suzhou 215123 Jiangsu Peoples R China Soochow Univ Sch Comp Sci & Technol Suzhou 215006 Peoples R China

Federated Learning (FL) is a distributed model training framework that allows multiple clients to collaborate on training a global model without disclosing their local data in edge computing (EC) environments. However, FL usually faces statistical heterogeneity (e.g., non-IID data) and system heterogeneity (e.g., computing and communication capabilities), resulting in poor model training performance. To deal with the above two challenges, we propose an efficient FL framework, named FedBR, which integrates the idea of block-wise regularization and knowledge distillation (KD) into the pioneering FL algorithm FedAvg, for resource-constrained edge computing. Specifically, we first divide the model into multiple blocks according to the layer order of deep neural network (DNN). The server only sends some consecutive model blocks instead of an entire model to clients for communication efficiency. Then, the clients make use of knowledge distillation to absorb the knowledge of global model blocks to alleviate statistical heterogeneity during local training. We provide a theoretical convergence guarantee for FedBR and show that the convergence bound will decrease as the increasing number of model blocks sent by the server. Besides, since the increasing number of model blocks brings more computing and communication costs, we design a heuristic algorithm (GMBS) to determine the appropriate number of model blocks for clients according to their varied data distributions, computing, and communication capabilities. Extensive experimental results show that FedBR can reduce the bandwidth consumption by about 31%, and achieve an average accuracy improvement of around 5.6% compared with the baselines under heterogeneous settings.

关键词： Federated learning distributed databases edge computing heterogeneity regularization knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

Secure Federated Learning for Multi-Party Network Monitoring

引用

IEEE ACCESS 2024年 12卷 163262-163284页

作者： Lytvyn, Oleksandr Nguyen, Giang Slovak Univ Technol Bratislava Fac Informat & Informat Technol Bratislava 84216 Slovakia Slovak Acad Sci Inst Informat Bratislava 84507 Slovakia

Network monitoring is essential for IT infrastructure health, enabling proactive threat detection, bandwidth optimization, and data analysis. Federated Learning facilitates collaboration and expands threat detection capabilities by allowing multiple clients to train models while preserving privacy and security without compromising sensitive information. This work explores the application of Federated Learning for secure network monitoring by investigating its use in various data partitioning settings to train Deep Learning models across multiple data partitions, incorporating secure aggregation to enhance data privacy. We propose an approach for multiparty training of deep neural networks for time series-based network load forecasting with secure aggregation. Our approach shows that a collaboratively trained model, on horizontal partitions, performs 11%-14% better than model training only on a single partition of that type. The model trained on vertical partitions achieved performance comparable to that of the models trained on a complete data set. Finally, examining the proposed approach on horizontal and vertical partitions proved its viability in both aggregation settings, regular and secure. These contributions demonstrate the feasibility of Federated Learning to improve interoperability between multiple organizations while addressing privacy and security concerns.

关键词： Monitoring Data models Federated learning Computational modeling Training Security Predictive models Load modeling distributed databases Context modeling Network monitoring federated learning secret sharing secure aggregation time-series modeling

来源：评论

学校读者我要写书评

暂无评论

Hybrid Fault-Tolerant and Attack-Resilient Cooperative Control in an Offshore Wind Farm

引用

IEEE TRANSACTIONS ON SUSTAINABLE ENERGY 2024年第2期15卷 1365-1379页

作者： Jadidi, Saeedreza Badihi, Hamed Zhang, Youmin Univ Saskatchewan Dept Elect & Comp Engn Saskatoon SK S7N 5A9 Canada Tampere Univ Fac Engn & Nat Sci Automat Technol & Mech Engn Unit Tampere 33720 Finland Concordia Univ Dept Mech Ind & Aerosp Engn Montreal PQ H3G 1M8 Canada

Modern wind farms promise increased capacity, sustainability, and efficiency through the integration of information technology with the existing wind energy technologies. However, this integration creates a new category of cyber-physical vulnerabilities stemming from physical faults and/or cyberattacks potentially leading to devastating physical impacts and catastrophic consequences. The large scale and high complexity of a wind farm, in addition to its growing connectivity, control functionality, and wind intermittency and variability make the task of cyber-physical protection particularly challenging. This paper introduces novel approaches for guaranteeing the safety, security, and reliability of a modern wind farm under simultaneous occurrence of faults and attacks using an advanced cyber-physical health monitoring scheme, defined as "intrusion detection and fault diagnosis system", as well as fault-tolerant/attack-resilient control strategies at different levels. The proposed fault-tolerant control strategy is based on adaptive model predictive control at turbine level, enhanced with a control reallocation mechanism at farm level. The attack-resilient control strategy is based on an automatic signal correction (ASC) technique that is applied at network operator level. The effectiveness of the suggested approaches is demonstrated using an offshore wind farm model under wind turbulences, measurement noises, and realistic physical fault and cyberattack scenarios.

关键词： Computer crime distributed databases Fault tolerant systems Fault tolerance Wind farms Reliability Security Attack-resilient control control reallocation cyberattack fault diagnosis fault-tolerant control intrusion detection model predictive control wind turbines wind farm

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Data Market Based on Multiagent Systems

引用

IEEE INTERNET COMPUTING 2024年第4期28卷 7-13页

作者： Baghcheband, Hajar Soares, Carlos Reis, Luis Paulo Univ Porto Fac Engn P-4200465 Porto Portugal Univ Porto Fac Engn Dept Comp Engn P-4200465 Porto Portugal

Today, autonomous agents, the Internet of Things, and smart devices produce more and more distributed data and use them to learn models for different purposes. One challenge is that learning from local data only may lead to suboptimal models. Thus, better models are expected if agents can exchange data, leading to approaches such as federated learning. However, these approaches assume that data have no value and, thus, is exchanged for free. A machine learning data market (MLDM), a framework based on multiagent systems with a market-based perspective on data exchange, was recently proposed. In an MLDM, each agent trains its model based on both local data and data bought from other agents. Although the empirical results are interesting, several challenges are still open, including data acquisition and data valuation. The MLDM is an illustrative example of how the value of data can and should be integrated into the design of distributed ML systems.

关键词： Data models distributed databases Cost accounting Biological system modeling Training Machine learning Computational modeling Autonomous agents Data analysis Market research Internet of Things Multi-agent systems Smart devices Information exchange

来源：评论

学校读者我要写书评

暂无评论

Cooperative Output Regulation of Unknown Linear Multiagent Systems: When Deadbeat Control Meets Data-Driven Framework

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2024年第5期20卷 7556-7564页

作者： Tian, Engang Zhai, Ganghui Liang, Dong Liu, Jinliang Univ Shanghai Sci & Technol Sch Opt Elect & Comp Engn Shanghai 200093 Peoples R China Univ Shanghai Sci & Technol Coll Sci Shanghai 200093 Peoples R China Nanjing Univ Finance & Econ Coll Informat Engn Nanjing 210023 Peoples R China Collaborat Innovat Ctr Modern Grain Circulat & Saf Nanjing 210023 Peoples R China

In this article, the cooperative output regulation problem (CORP) for unknown linear multiagent systems (MASs) is studied on the basis of a data-driven approach. Different from existing literature, first, the proposed method does not require prior knowledge of the system matrices and the leader system where the control gains are designed using input and state data. Next, the solution to the unknown regulator equations is derived using the data informativity approach. Then, to speed up the convergence rate, a distributed deadbeat control law is proposed for the MASs. For the purpose, the leader's system matrix is identified according to the measured data, and then a distributed observer is employed to estimate the state of the leader system in a distributed manner. Furthermore, the CORP is converted to a stabilization problem by using the data-based solution of the regulator equations. On the basis of the distributed observer, a data-driven deadbeat control law is designed in the context where the system matrices of the plant are completely unknown. It is guaranteed that the tracking error for each agent can converge to the origin asymptotically. Finally, an example is conducted to illustrate the effectiveness of the proposed data-based framework.

关键词： Mathematical models Observers Decentralized control Regulators distributed databases Regulation Data models Cooperative output regulation data-driven control multiagent systems (MASs)

来源：评论

学校读者我要写书评

暂无评论

Modeling and Performance Analysis of Multiserver Cloud Database Over Quasi-Static Rayleigh Fading Channel

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第11期11卷 20314-20326页

作者： Chen, Mengying Liu, Yang Dong, Chen An, Wannian Xu, Xiaodong Zhang, Ping Beijing Univ Posts & Telecommun Sch Informat & Commun Engn Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China

With the development of communication in the post-5G era, the combination of communication and cloud computing becomes closer. In order to promote the further development of cloud-network, this article will study the performance of Multiserver cloud Database under the Communication quality of quasi-static Rayleigh fading channel with multiple antennas (MC-MD). The CLIENTS with unlimited customers, a COMMUNICATION SYSTEM subject to quasi-static Rayleigh fading, and CLOUD DATABASE with two-phase locking protocol are the three components of the MC-MD model. Transactions are 1)initiated by the CLIENTS, 2)transmitted to the CLOUD DATABASE through the COMMUNICATION SYSTEM for processing, 3)then returned to the CLIENTS. The indicators of the model is mathematically derived by using queuing theory. These include client's indicators [average concurrent quantity of the system in steady state (CQ), average transactions stay time of the system in steady state (ST), average queue length of the waiting area in steady state (QL), and average transactions wait time of the waiting area in steady state (WT)] and server's indicator [average number of service desks in the busy period at steady state (DN)]. Under the appropriate conditions, the results indicate that the theoretical value of service performance is basically consistent with the simulation value. Clearly, the high speed improves the service performance of the system and decreases the service pressure. On the basis of this, the optimization strategy is proposed and the simulation indicators Jitter of transactions sojourn time of the system in steady state (STJ) is added. The results show that the transaction scheduling optimization strategy effectively reduce the delay and its jitter.

关键词： databases Cloud computing Protocols distributed databases Analytical models Computational modeling Servers Exclusive lock multiserver cloud database quasi-static Rayleigh fading channel with multiple antennas queue system shared lock

来源：评论

学校读者我要写书评

暂无评论

Locality-Preserving Graph Traversal With Split Live Migration

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2024年第10期35卷 1810-1825页

作者： Chen, Rong Wei, Xingda Xiating, Xie Chen, Haibo Shanghai Jiao Tong Univ Inst Parallel & Distributed Syst Shanghai 200240 Peoples R China Engn Res Ctr Domain Specif Operating Syst MOE Shanghai 200240 Peoples R China Chinese Acad Sci Key Lab Syst Software Beijing 100045 Peoples R China

Graph models many real-world data like social, transportation, biology, and communication data. Hence, graph traversal including multi-hop or graph-walking queries has been the key operation atop graph stores. However, since different graph traversals may touch different sets of vertices, it is hard or even impossible to have a one-size-fits-all graph partitioning algorithm that preserves access locality for various graph traversal workloads. Meanwhile, prior shard-based migration faces a dilemma such that coarse-grained migration may incur more migration overhead over increased locality benefits, while fine-grained migration usually requires excessive metadata and incurs non-trivial maintenance costs. We present Pragh, an efficient locality-preserving live graph migration scheme for graph stores in the form of key-value pairs. The key idea of Pragh is a split migration model that only migrates values physically while retaining keys in the initial location. This allows fine-grained migration while avoiding the need to maintain excessive metadata. Pragh integrates an RDMA-friendly location cache from DrTM-KV to provide fully-localized access to migrated data and further makes a novel reuse of the cache replacement policy for lightweight monitoring. Pragh further supports evolving graphs through a check-and-forward mechanism to resolve the conflict between updates and migration of graph data. Evaluations on an 8-node RDMA-capable cluster (100 Gbps) using a representative graph traversal benchmark show that Pragh can increase the throughput by up to 19x and decrease the median latency by up to 94%, thanks to split live migration that eliminates 97% remote accesses. A port of split live migration to Wukong shows up to 2.53x throughput improvement on representative workloads like LUBM-10240, thanks to a reduction of 88% remote accesses. This further confirms the effectiveness and generality of Pragh. Finally, though Pragh focuses on RDMA-based graph traversal, we show it

关键词： Throughput distributed databases Metadata Monitoring Social networking (online) Protocols Production Graph query graph traversal live migration locality preserving RDMA split migration

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：