检索结果-内蒙古大学图书馆

Accuracy evaluation of deep belief networks with fixed-point arithmetic

Computer Modelling and New Technologies 2014年第6期18卷 7-14页

作者： Jiang, Jingfei Hu, Rongdong Mikel, Lujan Dou, Yong Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology ChangSha Hunan China University of Manchester Manchester United Kingdom

Deep Belief Networks (DBNs) are state-of-art Machine Learning techniques and one of the most important unsupervised learning algorithms. Training DBNs is computationally intensive which naturally leads to investigate FPGA acceleration. Fixed-point arithmetic can be used when implementing DBNs in FPGAs to reduce execution time, but it is not clear the implications for accuracy. Previous studies have focused only on accelerators using some fixed bit-widths. A contribution of this paper is to demonstrate the bit-width effect on various configurations of DBNs in a comprehensive way by experimental evaluation. Explicit performance changing points are found using various bit-widths. The impact of sigmoid function approximation, required part of DBNs, is evaluated. A solution of mixed bit-widths DBN is proposed, fitting the bit-widths of FPGA primitives and gaining similar performance to the software implementation. Our results provide a guide to inform the design choices on bit-widths when implementing DBNs in FPGAs documenting clearly the trade-off in accuracy.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

A low-latency fine-grained dynamic shared cache management scheme for chip multi-processor

A low-latency fine-grained dynamic shared cache management s...

引用

IEEE International Conference on Performance, Computing and Communications (IPCCC)

作者： Jinbo Xu Weixia Xu Zhengbin Pang 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

In order to utilize the shared last-level cache (LLC) in chip multi-processors (CMP) more efficiently, the partitioning of LLC resources among all cores should have the characteristics of low-latency for access, fine granularity for migration and simple hardware complexity for implementation. This paper proposes a dynamic LLC management scheme to achieve these goals. The proposed scheme migrates cache resources among different cores at the granularity of cache blocks, instead of ways. The quantity of victim cache blocks that each victim core can migrate to other target cores are related to an eviction probability, which are calculated according to the performance goal. Then the victim cache blocks for a target core is chosen from the nearest victim core who has non-zero eviction probability by introducing innovate E-Table structure in CMP. The eviction probabilities are updated periodically. With the help of E-Tables, the proposal achieves low-latency accesses by always keeping the required cache blocks near to the target cores. And fine granularity is guaranteed by maintaining an eviction probability for each core. In addition, only little additional hardware changes to traditional cache structure is required. Simulation results suggest significant performance improvements from 6.8% to 22.7% over related works.

关键词： Hardware Resource management Probability distribution Proposals Complexity theory Simulation

来源：评论

学校读者我要写书评

暂无评论

ME-TCAM: Memory-Efficient Ternary Content Addressable Memory Based on Multipumping-Enabled LUTRAM on FPGA

ME-TCAM: Memory-Efficient Ternary Content Addressable Memory...

引用

Intelligent Computing and Human-Computer Interaction (ICHCI), International Conference on

作者： Zilin Shi Hui Yang Rulin Liu Jinli Yan Peng Qiao Baosheng Wang College of Computer Science and Technology National University of Defense Technology Changsha China Science and Technology on Parallel and Distributed Processing Laboratory (PDL) National University of Defense Technology Changsha China

Static random access memory (SRAM) on field programmable gate arrays (FPGAs) can be emulated to offer ternary content addressable memory (TCAM) functionality. However, SRAM-based TCAM wastes storage resources. This is due to the limited capacity of the physical addresses in the SRAM unit. This work proposes a LUTRAM-based TACM scheme on the FPGA called Memory-Efficient TCAM (ME-TCAM). METCAM divides SRAM unit into multiple virtual blocks mapping to a portion of the TCAM table to store the more address information of the TCAM table. Operation on SRAM block means that increasing the overall emulated TCAM bits/SRAM. Moreover, ME-TCAM exploits Xilinx primitives to conFigure lookup tables (LUTs) as 32 × 2 lookup table RAMs (LUTRAMs). We implement ME-TCAM using LUTRAM with a size of 512 × 48 and 1024 × 144 on a Virtex-7 FPGA device. Compared with the state-of-the-art research DUR, ME-TCAM achieves at least 2.6 times more memory efficiency.

关键词： Human computer interaction Associative memory Memory management Random access memory Logic gates Table lookup parallel architectures

来源：评论

学校读者我要写书评

暂无评论

Classification of Tiangong-1 hyperspectral remote sensing image via contextual sparse coding

Classification of Tiangong-1 hyperspectral remote sensing im...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Qi Lv Yong Dou Xin Niu Jiaqing Xu Jinbo Xu School 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

ISBN: (纸本)9781467372220

The hyperspectral remote sensing is one of the frontier techniques in the remote sensing research fields. Applying the sparse coding model to the hyperspectral remote sensing image processing is a hot topic in hyperspectral information processing. To improve the accuracy of hyperspectral image classification, we propose a classification method based on the spatial-spectral join-t contextual sparse coding. Firstly, a dictionary is obtained by training using samples selected from the ground-truth reference data. Then, the sparse coefficients of each pixel are calculated based on the learned dictionary. Afterward, the sparse coefficients are input to the classifier and the final classification result is obtained. The visible and near-infrared hyperspectral remote sensing image collected by Tiangong-1 in Chaoyang District of Beijing is used to evaluate the performance of the proposed approach. Experimental results show that the proposed method yields the best classification performance with the overall accuracy of 95.74% and the Kappa coefficient of 0.9476 in comparison with other classification methods.

关键词： Hyperspectral image Remote sensing Sparse coding Sparse coding Classification Methods remote sensing hyperspectral remote sensing hyperspectral imagery Image classification

来源：评论

学校读者我要写书评

暂无评论

Grid and Cooperative Computing - GCC 2004 1

引用

丛书名： Lecture Notes in Computer science

1000年

作者： Hai Jin Yi Pan Nong Xiao Jianhua Sun

ISBN: (数字)9783540302087

ISBN: (纸本)9783540235644

Welcome to the proceedings of GCC2004 and the city of Wuhan. Grid computing has become a mainstream research area in computer science and the GCC conference has become one of the premier forums for presentation of new and exciting research in all aspectsofgridandcooperativecomputing. Theprogramcommitteeispleasedtopresent the proceedings of the 3rd International Conference on Grid and Cooperative Comp- ing (GCC2004), which comprises a collection of excellent technical papers, posters, workshops, and keynote speeches. The papers accepted cover a wide range of exciting topics, including resource grid and service grid, information grid and knowledge grid, grid monitoring,managementand organizationtools, grid portal, grid service, Web s- vices and their QoS, service orchestration, grid middleware and toolkits, software glue technologies, grid security, innovative grid applications, advanced resource reservation andscheduling,performanceevaluationandmodeling,computer-supportedcooperative work, P2P computing, automatic computing, and meta-information management. The conference continues to grow and this year a record total of 581 manuscripts (including workshop submissions) were submitted for consideration. Expecting this growth, the size of the program committee was increased from 50 members for GCC 2003 for 70 in GCC 2004. Relevant differences from previous editions of the conf- ence: it is worth mentioning a signi?cant increase in the number of papers submitted by authors from outside China; and the acceptance rate was much lower than for p- vious GCC conferences. From the 427 papers submitted to the main conference, the program committee selected only 96 regular papers for oral presentation and 62 short papers for poster presentation in the program.

关键词： Theory of Computation Computer Communication Networks Operating Systems Software Engineering Information Systems Applications (incl. Internet) Information Storage and Retrieval

来源：评论

学校读者我要写书评

暂无评论

Synchronization Error Detection of MPI Programs by Symbolic Execution

Synchronization Error Detection of MPI Programs by Symbolic ...

引用

Asia-Pacific Conference on Software Engineering

作者： Xianjin Fu Zhenbang Chen Chun Huang Wei Dong Ji Wang 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

ISBN: (纸本)9781479974276

Asynchrony based overlapping of computation and communication is commonly used in MPI applications. However, this overlapping introduces synchronization errors frequently in asynchronous MPI programming. In this paper, we propose a symbolic execution based method for detecting input-related synchronization errors. The path space of an MPI program is systematically explored, and the related operations of the synchronization errors in the program are checked specifically. In addition, two optimizations are proposed to improve the efficiency. We have implemented our method as a prototype tool based on the symbolic executor Cloud9. The results of the extensive experiments indicate the effectiveness of our method.

关键词： Synchronization Data communication Libraries Optimization Buffer storage Prototypes Schedules

来源：评论

学校读者我要写书评

暂无评论

IDOpenFlow: An OpenFlow switch to support identifier-locator split communication

IDOpenFlow: An OpenFlow switch to support identifier-locator...

引用

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Yannan Yang Yaping Liu Zhihong Liu School 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

ISBN: (纸本)9781467386456

The traditional identifier locator split network has many issues such as inflexibility, hard to innovate and difficult to deploy. SDN (Software Defined Network) provides a new direction for designing flexible identifier locator split network. The recent identifier locator split network based on SDN use the OpenFlow switch directly via rewritting the address, which lacks the scalability and utilizes locator address ineffectively. An OpenFlow switch named IDOpenFlow is proposed to support the communication based on identifier. IDOpenFlow switch provides the communication mechanism via encapsulating the packets, which has good scalability and utilizing locator address effectively. IDOpenFlow switch encapsulates and decapsulates packets according flow entries which are installed by SDN controller. Moreover, the prototype system shows that IDOpenFlow effectively supports the communication for both the fixed node and the mobile node. With respect to the issues of software forwarding performance, a high-performance IDOpenFlow switch based on Intel DPDK (which is named A-IDOpenFlow) is proposed. The results of Ixia test tool show that: 1) for packets more than 128 bytes, A-IDOpenFlow switch supports the communication based on identifier at rate of 10Gbit/s; 2) for small packet of 64 bytes, the rate of A-IDOpenFlow is 7.25 times faster than the rate of IDOpenFlow.

关键词： Switches Protocols Software Routing Servers Scalability

来源：评论

学校读者我要写书评

暂无评论

Poster: Segmentation Based Online Performance Problem Diagnosis

Poster: Segmentation Based Online Performance Problem Diagno...

引用

International Conference on Software Engineering (ICSE)

作者： Jingwen Zhou Zhenbang Chen Ji Wang 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

ISBN: (纸本)9781479919352

Currently, the performance problems of software systems gets more and more attentions. Among various diagnosis methods based on system traces, principal component analysis (PCA) based methods are widely used due to the high accuracy of the diagnosis results and requiring no specific domain knowledge. However, according to our experiments, we have validated several shortcomings existed in PCA-based methods, including requiring traces with a same call sequence, inefficiency when the traces are long, and missing performance problems. To cope with these issues, we introduce a segmentation based online diagnosis method in this poster.

关键词： Principal component analysis Software systems Accuracy Measurement Monitoring Computer architecture Conferences

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Social Network Information Cascade Analysis: a survey

Deep Learning for Social Network Information Cascade Analysi...

引用

IEEE International Conference on Data science in Cyberspace (DSC)

作者： Liqun Gao Bin Zhou Yan Jia Hongkui Tu Ye Wang Chenguang Chen Haiyang Wang Hongwu Zhuang 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

The following topics are dealt with: learning (artificial intelligence); text analysis; natural language processing; pattern classification; graph theory; data mining; security of data; social networking (online); fea...

ISBN: (数字)9781728195582

ISBN: (纸本)9781728195599

关键词：

来源：评论

学校读者我要写书评

暂无评论

NR-MPI: A non-stop and fault resilient MPI supporting programmer defined data backup and restore for E-scale super computing systems

Supercomputing Frontiers and Innovations

引用

Supercomputing Frontiers and Innovations 2016年第1期3卷 4-21页

作者： Suo, Guang Lu, Yutong Liao, Xiangke Xie, Min Cao, Hongjia State Key Laboratory of High Performance Computing National University of Defense Technology Changsha Hunan Province China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha Hunan Province China

Fault resilience has became a major issue for HPC systems, particularly, in the perspective of future E-scale systems, which will consist of millions of CPU cores and other components. MPI-level fault tolerant constructs, such as ULFM, are being proposed to support software level fault tolerance. However, there are few systematic evaluations by application programmers using benchmarks or pseudo applications. This paper proposes NR-MPI, a Non-stop and Fault Resilient MPI, supporting programmer defined data backup and restore. To help programmers write fault tolerant programs, NR-MPI provides a set of friendly programming interfaces and a state transition diagram for data backup and restore. This paper focuses on design, implementation and evaluation of NR-MPI. Specifically,this paper puts emphases on failure detection in MPI library, friendly programming interface extending for NR-MPI and examples of fault tolerant programs based NRMPI. Furthermore, to support failure recovery of applications, NR-MPI implements data backup interfaces based on double in-memory checkpoint/*** conduct experiments with both NPB benchmarks and Sweep3D on TH supercomputer in NSCC-TJ. Experimental results show that NR-MPI based fault tolerant programs can recover from failures online without restarting, and the overhead is small even for applications with tens of thousands of cores. © The Authors 2016.

关键词： Message passing

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：