检索结果-内蒙古大学图书馆

51st Midwest Symposium on Circuits and Systems (MWSCAS 2008), vol.1

作者： Saumil G. Merchant Gregory D. Peterson NSF Center for High-Performance Reconfigurable Computing (NSF-CHREC) University of Florida Gainesville FL USA Department of Electrical Engineering and Computer Science University of Tennessee Knoxville TN USA

Dedicated hardware implementations of artificial neural networks promise to provide faster, lower power operation when compared to software implementations executing on processors. Unfortunately, custom hardware implementations do not support intrinsic training of these networks on-chip. Training is typically done using software simulations and the obtained network is synthesized and targeted to hardware offline. The ANN FPGA design presented here facilitates dynamic network structure and parameter changes required for intrinsic training of artificial neural networks, without reliance on runtime FPGA reconfigurations. This is an important feature for an online trainable system as typical reconfiguration cycle times on the order of a few milliseconds pose to be a major performance bottleneck for iterative training algorithms. The designed platform implements block-based neural networks and can be evolved and adapted in-field.

关键词： Field programmable gate arrays Artificial neural networks Hardware Neurons Training Biological neural networks Evolution (biology)

来源：评论

学校读者我要写书评

暂无评论

Overcoming Bandwidth and Satellite Communications Limitations to Accelerate Applications of Remote Sensing and high performance computing for Sustainable African Development: Contributions from Egypt, Nigeria and South Africa

Overcoming Bandwidth and Satellite Communications Limitation...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Gilbert L. Rochon Magdy Wahab Gamal El Afandi Ado Dan-Isa Happy Sithole Khomotso Kganyago Souleymane Fall Joseph Quansah Akilah Martin Bereket Araya Chetan Maringanti Cliff Robinson Kaiem L. Frink Jean Pierre Antelo Rosen Center for Advanced Computing Purdue Terrestrial Observatory Purdue University-Information Technology IN USA Department of Astronomy & Meteorology Faculty of Science Cairo University Cairo Egypt Department of Meteorology & Astronomy Al-Azhar University Cairo Egypt Department of Computer Engineering Bayero University Kano Nigeria Center for High Performance Computing CHCP-CSIRO Cape Town South Africa Department of Agricultural & Biological Engineering Purdue University West Lafayette IN USA School for New Learning De Paul University Chicago IL USA Purdue Research Opportunities Program West Lafayette IN USA

Sixty-two years after the first photos of earth taken by a camera aboard a V-2 rocket in October, 1946, and forty-eight years since the first CORONA satellite images were captured, the status of remote sensing research and applications within the African continent has made dramatic progress. Many African countries now have remote sensing research centers, within government agencies, research institutes and universities. Some countries in Africa currently have earth observing and/or telecommunications satellites in orbit and/or have such assets in various planning stages. The authors document such progress, in addition to the constraints to further applications of remote sensing for sustainable development in Africa, with special reference to data distribution constraints. Moreover, the authors address the urgency for bandwidth improvements within the African continent, so as to enable sustainable development initiatives to benefit from advances in high performance computing, required for ab initio near-real-time analysis of satellite-data. Such capabilities, it is argued, are propaedeutic for time-critical initiatives, such as vulnerability assessment, disaster preparedness and mitigation, emergency response, humanitarian assistance and post-calamity reconstruction, associated with a wide array of biogenic and anthropogenic disasters. Case studies of advances in infrastructure for satellite remote sensing and high performance computing, with implications for sustainable development in Africa, are provided from Egypt, Nigeria and South Africa.

关键词： Bandwidth Satellite communication Acceleration Remote sensing high performance computing Africa Sustainable development Earth Continents Cameras

来源：评论

学校读者我要写书评

暂无评论

On construction of a BioGrid platform for parallel bioinformatics applications

Proceedings of the ACM Symposium on Applied Computing

引用

Proceedings of the ACM Symposium on Applied computing 2007年 144-145页

作者： Yang, Chao-Tung Han, Tsu-Fen Kan, Heng-Chuan High Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Southern Business Unit National Center for High-performance Computing Hsinshi Tainan County 74147 Taiwan

ISBN: (纸本)1595934804;9781595934802

In this paper, we report on implementing an experimental distributed computing application for bioinformatics consisting of basic high-performance computing environments (Grid and PC Cluster systems), multiple interfaces at user portals that provide useful graphical interfaces to enable biologists who are not IT specialists to benefit directly from the use of high-performance technology. Copyright 2007 ACM.

关键词： Bioinformatics

来源：评论

学校读者我要写书评

暂无评论

A Real-time Relaxed Garbage Collection in DSM

引用

电子学报(英文版) 2007年第1期16卷 7-12页

作者： WU Junmin SONG Zhenhua Anhui Province Co-Key Laboratory of High Performance Computing and Application Department of Computer Science ＆ Technology University of Science ＆ Technology of China Hefei 230027 China

Garbage collection is a memory management mechanism for automatically reclaiming garbage objects in memory. It can effectively relieve the programmers' burden and optimize the design of programs. Currently many garbage collection methods for distributed systems have been put forth as the distributed systems are more and more popular. Relaxed consistency model is a very important kind of memory consistency model in distributed system, but it is inefficiently to perform garbage collection in the Distributed shared memory system (DSM) on relaxed memory model with traditional methods. This article proposes a trace-based garbage collection algorithm for DSM system based on relaxed memory model. Our algorithm notes the references among active objects of every node in DSM system, which makes the application process and the collection process able to execute concurrently. And it makes use of the relaxed consistency model's features to reduce communication cost and response delay. Furthermore, the use of two tables insures the algorithm correct and effective. This algorithm displays its advantages of real-time and incremental, and it is proved to be correct and effective.

关键词：垃圾回收算法分布式共享存储系统实时弱一致性模型 DSM

来源：评论

学校读者我要写书评

暂无评论

An overview of the haplotype problems and algorithms

引用

中国高等学校学术文摘·计算机科学 2007年第3期1卷 272-282页

作者： ZHAO Yuzhong XU Yun ZHANG Qiangfeng CHEN Guoliang Computer Science Department University of Science and Technology of ChinaHefei 230027China Anhui Province-MOST Co-Key Laboratory of High Performance Computing and Its Application Hefei 230027China

A single nucleotide polymorphism (SNP),as the most common form of genetic variation,has been widely studied to help analyze the possible association between diseases and *** gain more information,SNPs on a single chromosome are usually studied together,which constitute a *** haplotypes from biological experiments is usually very costly and time-consuming,which causes people to develop efficient methods to determine haplotypes from the computational *** problems and algorithms about haplotypes have been proposed to reduce the cost of studies of disease *** general,four categories of problems are widely researched:the haplotype assembly problem,the haplotype inference problem,the haplotype block partition problem,and the haplotype tagging SNP selection *** former two problems have been well reviewed by many researchers,whereas the latter two have not been comprehensively surveyed to our *** this paper,we try to make a detailed introduction to the four problems,especially the latter two.

关键词： bioinformatics haplotype single nucleotide polymorphism (SNP) algorithm complexity

来源：评论

学校读者我要写书评

暂无评论

Semantics-aware advanced OLAP visualization of multidimensional data cubes

引用

International Journal of Data Warehousing and Mining 2007年第4期3卷 1-30页

作者： Cuzzocrea, Alfredo Sacca, Domenico Serafino, Paolo Department of Electronics Computer Science and Systems University of Calabria Calabria Italy Institute of High Performance Computing and Networking Italian National Research Council Department of Computer Engineering University of Calabria Calabria Italy

Efficiently supporting advanced OLAP visualization of multidimensional data cubes is a novel and challenging research topic, which results to be of interest for a large family of data warehouse applications relying on the management of spatio-temporal (e.g., mobile) data, scientific and statistical data, sensor network data, biological data, etc. On the other hand, the issue of visualizing multidimensional data domains has been quite neglected from the research community, since it does not belong to the well-founded conceptual-logical-physical design hierarchy inherited from relational database methodologies. Inspired from these considerations, in this article we propose an innovative advanced OLAP visualization technique that meaningfully combines (i) the so-called OLAP dimension flattening process, which allows us to extract two-dimensional OLAP viewsfrom multidimensional data cubes, and (ii) very efficient data compression techniques for such views, which allow us to generate "semantics-aware" compressed representations where data are grouped along OLAP hierarchies. Copyright © 2007, IGI Global.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Models of parallel computation :a survey and classification

引用

中国高等学校学术文摘·计算机科学 2007年第2期1卷 156-165页

作者： ZHANG Yunquan CHEN Guoliang SUN Guangzhong MIAO Qiankun Laboratory of Parallel Computing Institute of SoftwareChinese Academy of SciencesBeijing 100080China State Key Laboratory of Computer Science Institute of SoftwareChinese Academy of SciencesBeijing 100080China Anhui Province-MOST Key Co-Lab of High Performance Computing and Its Applications Department of Computer Science and TechnologyUniversity of Science and Technology of ChinaHefei 230027China

In this paper,the state-of-the-art parallel computational model research is *** will introduce various models that were developed during the past *** to their targeting architecture features,especially memory organization,we classify these parallel computational models into three *** models and their characteristics are discussed based on three generations *** believe that with the ever increasing speed gap between the CPU and memory systems,incorporating non-uniform memory hierarchy into computational models will become *** the emergence of multi-core CPUs,the parallelism hierarchy of current computing platforms becomes more and more *** this complicated parallelism hierarchy in future computational models becomes more and more important.A semi-automatic toolkit that can extract model parameters and their values on real computers can reduce the model analysis complexity,thus allowing more complicated models with more parameters to be *** memory and hierarchical parallelism will be two very important features that should be considered in future model design and research.

关键词： parallel computational models hierarchicalmemory hierarchical parallelism three generations memorymodel

来源：评论

学校读者我要写书评

暂无评论

MPUS: A scalable parallel simulator for RedNeurons parallel computer (work-in-progress) 2

MPUS: A scalable parallel simulator for RedNeurons parallel ...

引用

2nd International Conference on Scalable Information Systems, InfoScale 2007

作者： Hui, Li Junming, Wu Guoliang, Chen Xiufeng, Sui Department of Computer Science and Technology Anhui Province Co-Key Laboratory of High Performance Computing and Application University of Science and Technology of China China

ISBN: (纸本)9781595937575

In this paper, we present a scalable parallel simulator - MPUS - for verifying the design of our next generation high performance parallel computer - RedNeurons(RN) parallel computer. The RedNeurons parallel computer is based on CMP technology, and it adopts an advanced but maybe some complicated architecture and topology. This paper mainly describes the design and implementation of the MPUS.

关键词： Simulators

来源：评论

学校读者我要写书评

暂无评论

Energy efficiency rate selection on a path in IEEE 802.11-based multi-hop networks

引用

Journal of Information and Computational science 2007年第2期4卷 757-766页

作者： Wang, Qingshan Xu, Yinlong Xu, Chenguang Chen, Guoliang Department of Computer Science and Technology University of Science and Technology of China Hefei 230027 China School of Sciences Hefei University of Technology Hefei 230009 China Key Laboratory of High Performance Computing and Its Applications Hefei 230027 China

Energy efficiency is one of the most important issues in IEEE 802.11-based multi-hop networks. In this paper we study the transmission rate selection on a path in the networks. The objective is to minimize the total energy consumption over the entire path subject to delay constraint. We formulate it as a multiple-choice knapsack problem, which has been proved to be NP-hard, and present an adaptive greedy rate selection algorithm (AGRSA) to get the near-optimal answer. The simulation results show that AGRSA can reduce more than 20% energy consumption than the classic technique that all nodes transmit packets with the highest rate.

关键词： Wireless networks

来源：评论

学校读者我要写书评

暂无评论

Parallel loop scheduling using knowledge-based workload estimation on grid environments

Parallel loop scheduling using knowledge-based workload esti...

引用

2007 International Symposium on Applications and the Internet, SAINT

作者： Shih, Wen-Chung Yang, Chao-Tung Chen, Chun-Jen Tseng, Shian-Shyong Department of Computer Science National Chiao Tung University Hsinchu 30010 Taiwan High-Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Department of Information Science and Applications Asia University Taichung 41354 Taiwan

ISBN: (纸本)0769527566

Parallel loop scheduling on grid environments is a challenging problem, especially for loops with irregular workload distribution. In the past, this problem of load imbalance resulting from irregular workload was not explicitly addressed. This paper proposes a new approach to schedule loop iterations with irregular workload on grid environments. Based on knowledge-based estimation of workload, the proposed method can dispatch an appropriate proportion of workload to each node for execution according to its performance. In addition, the scheduler uses historical statistics of CPU usage and network bandwidth to estimate the dynamically changing performance of each node. Two applications, regular type and irregular one respectively, are implemented and executed on a grid test-bed, which consists of four schools. Experimental results show that the new approach improves the performance on previous schemes. © 2007 IEEE.

关键词： Grid computing

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：