检索结果-内蒙古大学图书馆

Development and Prospects of the PARCS-WCF system

CYBERNETICS AND systemS ANALYSIS 2020年第1期56卷 152-158页

作者： Anisimov, A., V Fedorus, O. M. Taras Shevchenko Natl Univ Kyiv Kiev Ukraine

The paper considers details of implementation of the PARCS-WCF system. The PARCS-WCF system provides creation and operation of asynchronous-recursive processes in solving parallel computing tasks both on a single PC and on a computer network. It allows different modules to exchange information providing data exchange in systems with distributed and non-distributed memory.

关键词： PARCS-WCF PARCS parallel computing distributed computing system

来源：评论

学校读者我要写书评

暂无评论

= Dynamic Tasks Scheduling Model for Performance Evaluation of a distributed computing system through Artificial Neural Network

= Dynamic Tasks Scheduling Model for Performance Evaluation ...

引用

International Conference on Soft computing for Problem Solving (SocProS 2011)

作者： Singh, M. P. Yadav, P. K. Kumar, Harendra Agarwal, Babita Gurkula Kangari Univ Dept Math & Stat Hardwar 249404 Uttarakhand India Cent Bldg Res Inst Res Planing & Business Dev Roorkee 247667 Uttar Pradesh India

ISBN: (纸本)9788132204862

As technology has quickly and relentlessly advanced in the field of computer hardware, distributed computing system [DCS] has become increasingly popular. Performance enhancement is one of the most important issues in distributed systems. In this paper we have proposed a dynamic task allocation model based on artificial neural network [ANN] scheduling approach to arrange the tasks to the processors. Relocating the tasks from one processor to another at certain points during the course of execution of the program that contributes to the total cost of the running program has been taken into account. Phase-wise Execution Cost [EC], Inter Task Communication Cost [ITCC], Residence Cost [RC] of each task on different processors and Relocation Cost [REC] for each task has been considered while preparing the model.

关键词： distributed computing system Artificial Neural Network Phase-wise Execution Inter-Tasks Communication Cost Execution Cost Relocation Cost

来源：评论

学校读者我要写书评

暂无评论

distributed-based Hierarchical Clustering system for Large-scale Semiconductor Wafers

Distributed-based Hierarchical Clustering System for Large-s...

引用

IEEE International Conference on Industrial Engineering and Engineering Management (IEEE IEEM)

作者： Lee, Seungchul Kim, Daeyoung BISTel Inc Res Seoul South Korea Gwangju Inst Sci & Technol Sch Elect Engn & Comp Sci Gwangju South Korea Seoul Natl Univ Ind Engn Seoul South Korea

ISBN: (纸本)9781538667866

In this paper, we propose a distributed-based Hierarchical Clustering system for Large-Scale Semiconductor Wafers (DHCSSW). By applying the big-data framework to existing hierarchical clustering algorithm, the proposed system makes it feasible to cluster large-scale wafers with up to 320,000 wafers. To verify the performance of our approach, we used simulated wafer maps. The experimental results show that our system outperformed the existing hierarchical clustering in processing large-scale wafers, suggesting that currently used hierarchical clustering is insufficient in analyzing large-scale wafer maps. In addition, some failure patterns, which the existing approach is not able to detect, can be found with the DHCSSW. We anticipate that the DHCSSW will contribute in identifying the failure patterns in large-scale semiconductor wafers.

关键词： Big data analytics distributed computing system Hierarchical clustering Semiconductor wafer map

来源：评论

学校读者我要写书评

暂无评论

Computation of Birnbaum's Importance Using Logic Differential Calculus 42

Computation of Birnbaum's Importance Using Logic Differentia...

引用

42nd International Conference on Telecommunications and Signal Processing (TSP)

作者： Kvassay, Miroslav Rusnak, Patrik Sedlacek, Peter Univ Zilina Fac Management Sci & Informat Zilina Slovakia

ISBN: (纸本)9781728118642

Reliability analysis of technical systems, such as telecommunication, computer, or distribution networks, is a complex problem, which includes a lot of steps. Very important step is evaluation of importance of the system components for operation of the system. For this task, special indices known as importance measures are used. These measures can be structure, time-independent or time-dependent. One of the most commonly used importance measures is Birnbaum's Importance (BI). The structure version of this measure is defined as a relative number of situations in which a failure of a specific system component causes that the system fails. The time-independent version of this measure agrees with the probability that a failure of a given component results in a failure of the system, while the time-dependent version expresses how this probability changes over time. BI is used in computation of many other importance measures. This indicates that development of a unified approach for its computation can be used to develop a comprehensive framework for evaluation of importance of system components. In this paper, we present how logic differential calculus can be used for development of such an approach, and how various BI measures can be used in reliability analysis of a distributed computing system.

关键词： distributed computing system importance analysis logic differential calculus reliability structure function

来源：评论

学校读者我要写书评

暂无评论

Sparse Gradient Compression for distributed SGD 24th

Sparse Gradient Compression for Distributed SGD

引用

24th International Conference on Database systems for Advanced Applications (DASFAA)

作者： Sun, Haobo Shao, Yingxia Jiang, Jiawei Cui, Bin Lei, Kai Xu, Yu Wang, Jiang Peking Univ Sch Elect & Comp Engn Shenzhen Peoples R China BUPT Beijing Key Lab Intelligent Telecommun Software & Beijing Peoples R China Peking Univ Ctr Data Sci Beijing Peoples R China Natl Engn Lab Big Data Anal & Applicat Beijing Peoples R China Tencent Inc Shenzhen Peoples R China

ISBN: (纸本)9783030185794;9783030185787

Communication bandwidth is a bottleneck in distributed machine learning, and limits the system scalability. The transmission of gradients often dominates the communication in distributed SGD. One promising technique is using the gradient compression to reduce the communication cost. Recently, many approaches have been developed for the deep neural networks. However, they still suffer from the high memory cost, slow convergence and serious staleness problems over sparse high-dimensional models. In this work, we propose Sparse Gradient Compression (SGC) to efficiently train both the sparse models and the deep neural networks. SGC uses momentum approximation to reduce the memory cost with negligible accuracy degradation. Then it improves the accuracy with long-term gradient compensation, which maintains global momentum to make up for the information loss caused by the approximation. Finally, to alleviate the staleness problem, SGC updates model weight with the accumulation of delayed gradients at local, called local update technique. The experiments over the sparse high-dimensional models and deep neural networks indicate that SGC can compress 99.99% gradients for every iteration without performance degradation, and saves the communication cost up to 48x.

关键词： distributed computing system distributed optimization algorithm Gradient compression

来源：评论

学校读者我要写书评

暂无评论

The English Auction Method for Scheduling Jobs in a distributed Network of Supercomputer Centers

引用

LOBACHEVSKII JOURNAL OF MATHEMATICS 2019年第5期40卷 606-613页

作者： Tikhomirov, A. I. Russian Acad Sci Joint Supercomp Ctr Moscow 119334 Russia

The article describes the problem of scheduling jobs with absolute priorities in a geographically distributed network of supercomputer centers (GDN). In this case English auction method can be efficiently applied. Classic market model considers computational resources as the goods (subject of auction trades), and resources' owners act as sellers. Users act as buyers who participate in the auction on purpose to purchase computing resources for the execution of their jobs. This model assumes that customers have certain budgets in nominal or real money. The priority of the job is actually determined by the price, which the user can pay to finish the job by a certain time. The GDN model investigated by the authors differs from the known ones in that the jobs priorities are absolute and assigned according to uniform rules. The main goal is the earliest execution of high-priority jobs. In this case, the concept of the user's budget becomes meaningless, and the classic auction models do not work. The authors propose a new approach where the jobs act as the goods and buyers are resource owners who paying for jobs with available idle supercomputing resources. For this approach, the authors investigate the features and characteristics of English auction, as the most preferred method for scheduling jobs with absolute priorities in GDN.

关键词： high performance computing supercomputer center English auction distributed computing system

来源：评论

学校读者我要写书评

暂无评论

Constrained Multi-Objective Optimization for Automated Machine Learning 6

Constrained Multi-Objective Optimization for Automated Machi...

引用

6th IEEE International Conference on Data Science and Advanced Analytics (DSAA)

作者： Gardner, Steven Golovidov, Oleg Griffin, Joshua Koch, Patrick Thompson, Wayne Wujek, Brett Xu, Yan SAS Inst Inc Cary NC 27513 USA

ISBN: (纸本)9781728144931

Automated machine learning has gained a lot of attention recently. Building and selecting the right machine learning models is often a multi-objective optimization problem. General purpose machine learning software that simultaneously supports multiple objectives and constraints is scant, though the potential benefits are great. In this work, we present a framework called Autotune that effectively handles multiple objectives and constraints that arise in machine learning problems. Autotune is built on a suite of derivative-free optimization methods, and utilizes multi-level parallelism in a distributed computing environment for automatically training, scoring, and selecting good models. Incorporation of multiple objectives and constraints in the model exploration and selection process provides the flexibility needed to satisfy trade-offs necessary in practical machine learning applications. Experimental results from standard multi-objective optimization benchmark problems show that Autotune is very efficient in capturing Pareto fronts. These benchmark results also show how adding constraints can guide the search to more promising regions of the solution space, ultimately producing more desirable Pareto fronts. Results from two real-world case studies demonstrate the effectiveness of the constrained multi-objective optimization capability offered by Autotune.

关键词： Multi-objective Optimization Automated Machine Learning distributed computing system

来源：评论

学校读者我要写书评

暂无评论

Random ensemble learning for EEG classification

引用

ARTIFICIAL INTELLIGENCE IN MEDICINE 2018年 84卷 146-158页

作者： Hosseini, Mohammad-Parsa Pompili, Dario Elisevich, Kost Soltanian-Zadeh, Hamid Rutgers State Univ Dept Elect & Comp Engn Piscataway NJ 08854 USA Michigan State Univ Coll Human Med Div Neurosurg Grand Rapids MI 49503 USA Spectrum Hlth Dept Clin Neurosci Grand Rapids MI 49503 USA Univ Tehran Sch Elect & Comp Engn CIPCE Tehran Iran Henry Ford Hlth Syst Image Anal Lab Dept Radiol Detroit MI 48202 USA Henry Ford Hlth Syst Image Anal Lab Dept Res Adm Detroit MI 48202 USA

Real-time detection of seizure activity in epilepsy patients is critical in averting seizure activity and improving patients' quality of life. Accurate evaluation, presurgical assessment, seizure prevention, and emergency alerts all depend on the rapid detection of seizure onset. A new method of feature selection and classification for rapid and precise seizure detection is discussed wherein informative components of electroencephalogram (EEG)-derived data are extracted and an automatic method is presented using infinite independent component analysis (I-ICA) to select independent features. The feature space is divided into subspaces via random selection and multichannel support vector machines (SVMs) are used to classify these subspaces. The result of each classifier is then combined by majority voting to establish the final output. In addition, a random subspace ensemble using a combination of SVM, multilayer perceptron (MLP) neural network and an extended k-nearest neighbors (k-NN), called extended nearest neighbor (ENN), is developed for the EEG and electrocorticography (ECoG) big data problem. To evaluate the solution, a benchmark ECoG of eight patients with temporal and extratemporal epilepsy was implemented in a distributed computing framework as a multitier cloud-computing architecture. Using leave-one-out cross-validation, the accuracy, sensitivity, specificity, and both false positive and false negative ratios of the proposed method were found to be 0.97, 0.98, 0.96, 0.04, and 0.02, respectively. Application of the solution to cases under investigation with ECoG has also been effected to demonstrate its utility. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Brain-computer interface distributed computing system Electroencephalogram Ensemble learning Epileptic seizure detection Computational neuroscience

来源：评论

学校读者我要写书评

暂无评论

Autotune: A Derivative-free Optimization Framework for Hyperparameter Tuning 18

Autotune: A Derivative-free Optimization Framework for Hyper...

引用

24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)

作者： Koch, Patrick Golovidov, Oleg Gardner, Steven Wujek, Brett Griffin, Joshua Xu, Yan SAS Inst Inc Cary NC 27513 USA

ISBN: (纸本)9781450355520

Machine learning applications often require hyperparameter tuning. The hyperparameters usually drive both the efficiency of the model training process and the resulting model quality. For hyperparameter tuning, machine learning algorithms are complex black-boxes. This creates a class of challenging optimization problems, whose objective functions tend to be nonsmooth, discontinuous, unpredictably varying in computational expense, and include continuous, categorical, and/or integer variables. Further, function evaluations can fail for a variety of reasons including numerical difficulties or hardware failures. Additionally, not all hyperparameter value combinations are compatible, which creates so called hidden constraints. Robust and efficient optimization algorithms are needed for hyper-parameter tuning. In this paper we present an automated parallel derivative-free optimization framework called Autotune, which combines a number of specialized sampling and search methods that are very effective in tuning machine learning models despite these challenges. Autotune provides significantly improved models over using default hyperparameter settings with minimal user interaction on real-world applications. Given the inherent expense of training numerous candidate models, we demonstrate the effectiveness of Autotune's search methods and the efficient distributed and parallel paradigms for training and tuning models, and also discuss the resource trade-offs associated with the ability to both distribute the training process and parallelize the tuning process.

关键词： Derivative-free Optimization Stochastic Optimization Bayesian Optimization Hyperparameters distributed computing system

来源：评论

学校读者我要写书评

暂无评论

The Economics of Introducing the Peer-to-peer system of Storage and Processing of Protected Information at an Enterprise

The Economics of Introducing the Peer-to-peer System of Stor...

引用

International Conference on Complex systems: Control and Modeling Problems

作者： Alexey Nenashev Victor Khryashchev Samara State Technical University

ISBN: (纸本)9781728167015

The method of evaluating the effectiveness of the economic parameters for the implementation at industrial enterprises of "Peer-to-Peer distributed Storage and Processing of Protected Information in Corporate Networks," designed primarily to protect distributed computing and isolate private networks without limiting the possibilities of useful interaction in the framework of the technologies Industry 5.0 and Super Smart Society is proposed. Within the framework of these concepts, there is a gradual integration of production, logistics, sales, and information systems into a global information exchange system. The proposed system provides for effective counteraction to the threats using cryptography, protection against unauthorized access using biometrics, and an innovative protocol for information interaction and control of topology based on distributed ledger technologies, it optimizes the use of computing power and provides tools for distributed computing without changing the logic of existing software.

关键词： Industry 5.0 Super smart society distributed ledger Cybersecurity Performance evaluation Computer simulation Peer-to-peer networks Information security distributed computing systems distributed computing system Network security peer to peer file sharing computer information security Performance Evaluation Computer Simulation distributed computing economics Peer-to-peer network Confidential Information peer to peer Private networks Industrial enterprises unauthorized access

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：