检索结果-内蒙古大学图书馆

Optimization approaches of organizing streams on imagine processor

Jisuanji Xuebao/Chinese Journal of Computers 2008年第7期31卷 1092-1100页

作者： Yang, Xue-Jun Zeng, Li-Fang Deng, Yu Tang, Yu-Hua National Key Laboratory of Parallel and Distributed Processing School of Computer Science National University of Defense Technology Changsha 410073 China Jiangnan Remote Sensing Institute Shanghai 200072 China

Due to the characteristics of stream applications and the insufficiency of conventional processors when running stream programs, stream processors which support data-level parallelism become the research hotspot. This paper presents two means, stream partition (SP) and stream compression (SC), to optimize streams on Imagine. The results of simulation show that SP and SC can make stream applications take full advantage of the parallel clusters, pipelines and three-level memory hierarchy of the Imagine processor, and then reduce the execution time of stream programs.

关键词： parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Trace Bench: An Open Data Set for Trace-Oriented Monitoring

Trace Bench: An Open Data Set for Trace-Oriented Monitoring

引用

IEEE International Conference on Cloud Computing Technology and Science (CloudCom)

作者： Jingwen Zhou Zhenbang Chen Ji Wang Zibin Zheng Michael R. Lyu College of Computer National University of Defense Technology Changsha China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha China Shenzhen Research Institute The Chinese University of Hong Kong Shenzhen China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong China

User request trace-oriented monitoring is an effective method to improve the reliability of cloud systems. However, there are some difficulties in getting traces in practice, which hinder the development of trace-oriented monitoring research. In this paper, we release a fine-grained user request-centric open trace data set, called Trace Bench, collected on a real world cloud storage system deployed in a real environment. During collecting, many aspects are considered to simulate different scenarios, including cluster size, request type, workload speed, etc. Besides recording the traces when the monitored system is running normally, we also collect the traces under the situation with faults injected. With a mature injection tool, 14 faults are introduced, including function faults and performance faults. The traces in Trace Bench are clustered in different files, where each file corresponds to a certain scenario. The whole collection work lasted for more than half a year, resulting in more than 360, 000 traces in 361 files. In addition, we also employ several applications based on Trace Bench, which validate the helpfulness of Trace Bench for the field of trace-oriented monitoring.

关键词： Monitoring Servers Educational institutions Instruments Cloud computing Context Fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

Symmetric Non-negative Matrix Factorization Based Link Partition Method for Overlapping Community Detection

Symmetric Non-negative Matrix Factorization Based Link Parti...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Xiang Zhang Naiyang Guan Wenju Zhang Xuhui Huang Shuyi Wu Zhigang Luo Science and Technology on Parallel and Distributed Processing Laboratory College of Computer National University of Defense Technology Hunan P.R. China Institute of Software College of Computer National University of Defense Technology Hunan P.R. China Department of Computer Science and Technology College of Computer National University of Defense Technology Hunan P.R. China

ISBN: (纸本)9781479986989

Partitioning links rather than nodes is effective in overlapping community detection (OCD) on complex networks. However, it consumes high CPU and memory overheads because the volume of links is huge especially when the network is rather complex. In this paper, we proposes a symmetric non-negative matrix factorization (SNMF) based link partition method called SNMF-Link to overcome this deficiency. In particular, SNMF-Link represents data in a lower-dimensional space spanned by the node-link incidence matrix. By solving a lighter SNMF problem, SNMF-Link learns the clustering indicators of each links. Since traditional multiplicative update rule (MUR) based optimization algorithm for SNMF suffers from slow convergence, we applied the augmented Lagrangian method (ALM) to efficiently optimize SNMF. Experimental results show that SNMF-Link is much more efficient than the representative clustering algorithms without reducing the OCD performance.

关键词： Symmetric matrices Convergence Optimization Partitioning algorithms Chlorine Complex networks Image edge detection

来源：评论

学校读者我要写书评

暂无评论

Correntropy induced metric based graph regularized non-negative matrix factorization

Correntropy induced metric based graph regularized non-negat...

引用

International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)

作者： Bin Mao Naiyang Guan Dacheng Tao Xuhui Huang Zhigang Luo College of Science National University of Defense Technology Changsha China College of Computer National University of Defense Technology Science and Technology on Parallel and Distributed Processing Laboratory Changsha China Department of Computer Science and Technology College of Computer National University of Defense Technology Changsha China Faculty of Engineering and Information Technology University of Technology Sydney Sydney NSW Australia

Non-negative matrix factorization (NMF) is an efficient dimension reduction method and plays an important role in many pattern recognition and computer vision tasks. However, conventional NMF methods are not robust since the objective functions are sensitive to outliers and do not consider the geometric structure in datasets. In this paper, we proposed a correntropy graph regularized NMF (CGNMF) to overcome the aforementioned problems. CGNMF maximizes the correntropy between data matrix and its reconstruction to filter out the noises of large magnitudes, and expects the coefficients to preserve the intrinsic geometric structure of data. We also proposed a modified version of our CGNMF which construct the adjacent graph by using sparse representation to enhance its reliability. Experimental results on popular image datasets confirm the effectiveness of CGNMF.

关键词： Robustness Vectors Clustering algorithms Mutual information Measurement Computer integrated manufacturing

来源：评论

学校读者我要写书评

暂无评论

Maximizing the information diffusion opportunity in the cyber-physical network

Maximizing the information diffusion opportunity in the cybe...

引用

International Conference on Quality of Service in Heterogeneous Wired/Wireless Networks (QSHINE)

作者： Hongliang Lu Xuan Dong Wenxiang Li Saohe Lv Xiaodong Wang Wei Chen Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Hunan China School of Information Science and Engineering Wuhan University of Science and Technology Wuhan China Engineering Research Center for Metallurgical Automation and Detecting Technology of Ministry of Education Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9781479982172

Our daily life is changing by the smart objects, such as smart watches, smart phones etc. They make the cyber world and the physical world integrated by their abundant abilities of sensing, communication and computation etc. Focusing on a wide range of the integrated network, a statistical based strategy was introduced to get a special kind of link between objects, the statistical probability communication link. To get a maximized information spread probability for grouped people, this paper introduced a distributed, yet efficient algorithm naming DMPID algorithm, for finding a sub-network to spread people oriented inforamtion. The DMPID algorithm take the size of the selection and the information spread probability into account, and made a balance between the two parameters. Extended simulation showed that the DMPID algorithm performs well in different distributed networks.

关键词： distributed algorithms probability statistical analysis telecommunication networks DMPID algorithm cyber world cyber-physical network distributed maximizing probability of information diffusion information diffusion opportunity information spread probability statistical probability communication link

来源：评论

学校读者我要写书评

暂无评论

Scalable unbiased sampling method based on multi-peer adaptive random walk

引用

Ruan Jian Xue Bao/Journal of Software 2009年第3期20卷 630-643页

作者： Fu, Yong-Quan Wang, Yi-Jie Zhou, Jing National Key Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China Staff Room of Computer Command Academy of the Corps of Engineers Xuzhou 221000 China

To deal with the scalable and fast unbiased sampling problems in unstructured P2P systems, a sampling method based on multi-peer adaptive random walk (SMARW) is proposed. In the method, based on the multi-peer random walk process, a set of provisional peers are selected as agents which start the sampling processes, by which the sampling process is speeded up with receiving a set of tunable number samples each time;Meanwhile, after receiving new samples earlier agents are replaced with these new samples which repeat the sampling process. With this simple replacement, it can be guaranteed with high probability that the system can reach the optimal load balance;furthermore, SMARW adopts an adaptive distributed random walk adjustment process to increase the convergence rate of the sampling process. A detailed theorical analysis and performance evaluation confirm that SMARW has a high level of unbiased sampling and near-optimal load balancing capability. © by Institute of Software, the Chinese Academy of Sciences. All rights reserved.

关键词： Peer to peer networks

来源：评论

学校读者我要写书评

暂无评论

Deep reinforcement learning for combinatorial optimization: Covering salesman problems

arXiv

引用

arXiv 2021年

作者： Li, Kaiwen Zhang, Tao Wang, Rui Wang, Yuheng Han, Yi College of Systems Engineering National University of Defense Technology Changsha410073 China Hunan Key Laboratory of Multi-Energy System Intelligent Interconnection Technology HKL-MSI2T Changsha410073 China Graduate College National University of Defense Technology Changsha410073 China Science and Technology on Parallel and Distributed Processing Laboratory College of Computer National University of Defense Technology Changsha410073 China

This paper introduces a new deep learning approach to approximately solve the Covering Salesman Problem (CSP). In this approach, given the city locations of a CSP as input, a deep neural network model is designed to directly output the solution. It is trained using the deep reinforcement learning without supervision. Specifically, in the model, we apply the Multi-head Attention to capture the structural patterns, and design a dynamic embedding to handle the dynamic patterns of the problem. Once the model is trained, it can generalize to various types of CSP tasks (different sizes and topologies) with no need of re-training. Through controlled experiments, the proposed approach shows desirable time complexity: it runs more than 20 times faster than the traditional heuristic solvers with a tiny gap of optimality. Moreover, it significantly outperforms the current state-of-the-art deep learning approaches for combinatorial optimization in the aspect of both training and inference. In comparison with traditional solvers, this approach is highly desirable for most of the challenging tasks in practice that are usually large-scale and require quick decisions. Copyright © 2021, The Authors. All rights reserved.

关键词： Combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

Two layered P2P model for semantic service discovery

引用

Ruan Jian Xue Bao/Journal of Software 2007年第8期18卷 1922-1932页

作者： Liu, Zhi-Zhong Wang, Huai-Min Zhou, Bin Institute of Network and Information Security School of Computer National University of Defense Technology Changsha 410073 China National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China

In open Internet environment, it is inevitable that multiple ontologies coexist. Centralized service discovery mechanism becomes the bottleneck of SOC (service oriented computing), which results in poor scalability of system. Aiming at solving these problems, a two layered P2P based model for semantic service discovery is proposed in this paper. The model is based on ontology community and integrates iVCE (Internet-based virtual computing environment) core concepts into a P2P model. Based on this model, a service discovery algorithm composed of two stages and three steps is proposed. It matches services across communities as well as within community. Within a community, algorithm locates registers holding service information with a high probability of satisfying a request firstly. Then it captures semantic matching between service advertisements and service requests by logical reasoning. Service discovery across communities occurs according to some policies. The model is suitable for opening environment with coexistent multiple ontologies. Experimental results show that given an appropriate setting, the model can make a tradeoff between recall and responding time. In addition, the model will release the mean load of registers efficiently while holding recall.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Accelerating MatLab code using GPU: A review of tools and strategies

Accelerating MatLab code using GPU: A review of tools and st...

引用

International Conference on Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC)

作者： Baida Zhang Shuai Xu Feng Zhang Yuan Bi Linqi Huang National laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha China Department of Information Engineering Academy of Armored Forces Engineering Beijing China Inspur (Beijing) Electronic Information Industry Company Limited Beijing China Naval Academy of Armament Beijing China School of Resources and Safety Engineering Central South University Changsha China

Lots of toolboxes of accelerating MatLab using GPU are available now[1], but, users are confused by which toolbox is best suitable for a particular task. Three toolboxes-Jacket, GPUmat, and parallel Computing Toolbox of MatLab are selected. For each toolbox, its advantages and pitfalls are reviewed, with an aim to allow the reader to identify which toolbox is appropriate for a given task. Strategies of whether a function should execute on GPU are given after a formula analysis. The analysis is also a framework for program automatically decides which function is cost-efficient to execute on GPU. A series of benchmark of different types of computing, including data transfer between GPU and CPU, data matrix Generation, matrix operation and GPU functions were tested in all three toolboxes. And the results show that Jacket is the best one. Some advices to improve the performance of toolboxes are given in the end.

关键词： Graphics processing unit Acceleration parallel processing Computer architecture Conferences Licenses MATLAB

来源：评论

学校读者我要写书评

暂无评论

A GPU-based parallel WFST decoder on nnet3

引用

AIP Conference Proceedings 2019年第1期2073卷

作者： Yong Wang Jie Liu Chen Zhou Zhengbin Pang Shengguo Li Chunye Gong Xinbiao Gan Yurong Li 1Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha 410073 China

One performance-intensive part of automatic speech recognition is the weighted finite-state transducer (WFST) decoding. To solve the problem, we expand parallel Graphics processing Units (GPU) computing to the decoding period. We describe extension work based on Kaldi toolkit for speech recognition research. Our work can support weighted finite-state transducer decoding on Kaldi neural nets with CUDA toolkit. Our paper also expands an efficient parallel Viterbi beam decoding algorithm to decrease the speech recognition Real Time Factor (RTF) value. Together with our optimization algorithm, we have reached 2.3x speed up on the AISHELL corpus decoding. We also implement nnet3 decoder that improves real-time speed up with no word error rate raise.

关键词：

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：