检索结果-内蒙古大学图书馆

arXiv 2017年

作者： Kuehn, Martin Keuper, Janis Pfreundt, Franz-Josef Competence Center High Performance Computing Fraunhofer Institute for Industrial Mathematics Fraunhofer-Platz 1 KaiserslauternD-67663 Germany

deep Neural Network (dNN) are currently of great interest in research and application. The training of these networks is a compute intensive and time consuming task. To reduce training times to a bearable amount at reasonable cost we extend the popular Caffe toolbox for dNN with an efficient distributed memory communication pattern. To achieve good scalability we emphasize the overlap of computation and communication and prefer fine granular synchronization patterns over global barriers. To implement these communication patterns we rely on the the"Global address space Programming Interface" version 2 (GPI-2) communication library. This interface provides a light-weight set of asynchronous one-sided communication primitives supplemented by non-blocking fine granular data synchronization mechanisms. Therefore, CaffeGPI is the name of our parallel version of Caffe. First benchmarks demonstrate better scaling behavior compared with other extensions, e.g., the IntelTMCaffe. Even within a single symmetric multiprocessing machine with four graphics processing units, the CaffeGPI scales better than the standard Caffe toolbox. These first results demonstrate that the use of standard high performance computing (HPC) hardware is a valid cost saving approach to train large ddNs. I/O is an other bottleneck to work with ddNs in a standard parallel HPC setting, which we will consider in more detail in a forthcoming paper. Copyright © 2017, The Authors. All rights reserved.

关键词： deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Nanorattle Au@PtAg encapsulated in ZIF-8 for enhancing CO2 photoreduction to CO

引用

Nano research 2019年第3期12卷 625-630页

作者： Yuqun Su Haitao Xu Jiajia Wang Xikuo Luo Zhen-liang Xu Kefu Wang Wenzhong Wang State Key Laboratory of Chemical Engineering Membrane Science and Engineering R&D Lab Chemical Engineering Research Center East China University of Science and Technology (ECUST) 130 Meilong Road Shanghai 200237 China State Key Laboratory of High Performance Ceramics and Super Fine Microstructure Shanghai Institute of Ceramics Chinese Academy of Sciences 1295 Dingxi Road Shanghai 200050 China

Imidazolate-based ZIF-8 catalysts M@ZIF-8 (M = Au Nr, Au@Ag Nr, or Au@PtAg NrT;Nr = nanorod, NrT = nanorattle), were assembled. Au Nrs acted as the core for the epitaxial growth of the Ag shell, and oxidative etching of Au@Ag Nrs led to Au@PtAg NrTs with K2PQI4 aqueous soluti on. All metal nano rods (MNrs) and metal nano rattles (MNrTs) were well dispersed and fully en capsulated in ZIF-8. Au @ PtAg NrTs encapsulated in ZIF-8 could lead to enhanced stability and selectivity for catalytic applications, combining the advantages of ZIF-8 (tailorable porosity) with the high surface area and improved optical sensitivity of rod-shaped NrTs. The catalyst Au @ PtAg@ZIF-8 exhibited efficient catalytic activity and CO selectivity for the gas-phase photoreduction of CO2 with H2O.

关键词： imidazolate-based ZIF-8 core-shell nanorattle noble metal catalyst CO2 photoreduction

来源：评论

学校读者我要写书评

暂无评论

New driving method for triode CNT-FEd 16

New driving method for triode CNT-FED

引用

16th International Vacuum Microelectronics Conference, IVMC 2003

作者： Lee, Chun-Tao Lan, Yung-Chiang Tsui, Bing-Yue Lee, Cheng-Chung ERSO ITRI Sec. 4 195-4 Chung Hsing Rd. Chutung Hsinchu Taiwan National Center for High-performance Computing Science Park 7 R and D Rd. VI Hsinchu Taiwan Department of Electronics Engineering Institute of Electronics National Chiao-Tung University Hsinchu Taiwan

来源：评论

学校读者我要写书评

暂无评论

NETWOrKEd COLLABOrATIVE ENVIrONMENT FOr HYdrO-ENGINEErING

引用

International Journal of Computational Engineering Science 2001年第4期2卷 557-567页

作者： SHIN-JYE LIANG JIUNN-HOrNG LEE SHIH-HUNG CHEN Jr-HUANG SHIAU Department of Hydraulic Engineering Feng Chia University Taichung Taiwan R.O.C National Center for High-Performance Computing Hsinchu Taiwan R.O.C

Engineering projects are increasingly becoming more complex and have to be carried out in close cooperation among several experts from different disciplines and locations. This has resulted from a rapid progress in communication, computing, visualization, and storage technology. research and applications of distributed computing, visualization and networked collaboration, therefore, is becoming an important subject. We present three ongoing distributed collaborative systems at the National center for high-performance computing in Taiwan: (1) A collaborative visualization environment which supports distributed and interactive visualization and collaboration in the three-dimensional virtual world, (2) A collaborative web browser system that creates an HTML document shared environment for supporting remote conferencing, and (3) An Internet-based distributed analysis environment that is designed for distributed simulation and visualization. We have explored and examined various computer tools and communication technologies for the development of the networked collaborative systems. Our endeavor sheds some light on the future development and applications of an efficient networked collaborative infrastructure which supports distributed computing, presentation and visualization, data and information management, documentation, and conferencing.

关键词： information technology software integration networked analysis visualization and collaboration hydroinformatics

来源：评论

学校读者我要写书评

暂无评论

Simulation of 3d flow in a fish-bone-type fishway

Simulation of 3D flow in a fish-bone-type fishway

引用

3rd IASTEd International Conference on Environmental Modelling and Simulation, EMS 2007

作者： Chang, Wen-Yi Lee, Lung-Cheng Lien, Ho-Cheng Lin, Fang-Pang Tsai, Whey-Fong Lai, Jihn-Sung Tan, Yih-Chi National Center for High-Performance Computing No. 7 R and D Rd. VI Hsinchu Science Park Hsinchu 30076 Taiwan Hydrotech Research Institute Department of Bioenvironmental Systems Eng. National Taiwan University Taipei City 106 Taiwan WG-ENV ICID. SG Chinese Taipei Committee

ISBN: (纸本)9780889866829

A 3d two-phase flow numerical model based on the large eddy simulation is developed to deal with the free surface problems by using the surface capturing method. This numerical model is employed for simulations of 3d flow fields in a fish-bone-type fishway. The specific flow patterns induced by fish-bone-type obstructions at the bottom of the channel is illustrated. For assessment of fishway suitability, comparisons of flow fields at different swimming speeds of fishes are made to investigate the potential influences of the flow fields on the migration behaviors of fishes.

关键词： Fishways

来源：评论

学校读者我要写书评

暂无评论

Usage centric green performance indicators

Usage centric green performance indicators

引用

GreenMetrics Workshop 2011

作者： Chen, doron Henis, Ealan Kat, ronen I. Sotnikov, dmitry Cappiello, Cinzia Ferreira, Alexandre Mello Pernici, Barbara Vitali, Monica Jiang, Tao Liu, Jia Kipp, Alexander IBM Research Haifa University Campus Mount Carmel Haifa31905 Israel Politecnico di Milano Via G. Ponzio 34/5 Milan20133 Italy High Performance Computing Center Stuttgart Nobelstr. 19 StuttgartD-70569 Germany

Energy efficiency of data centers is gaining importance as energy consumption and carbon footprint awareness are rising. Green performance Indicators (GPIs) provide measurable means to assess the energy efficiency of a resource or system. Most of the metrics commonly used today measure the energy efficiency potential of a resource, system or application usage, rather than the energy efficiency of the actual usage. In this paper, we argue that the way that the resources and systems are actually used in a given data center configuration is at least as important as the efficiency potential of the raw resources or systems. Hence, for data center energy efficiency, we suggest to both select energy efficient components (as done today), as well as optimize the actual usage of the components and systems in the data center. To achieve the latter, optimization of usage centric GPI metrics should be employed and targeted as a primary green goal. In this paper we identify and present usage centric metrics, which should be monitored and optimized for improving energy efficiency, and hence, reduce the data center carbon footprint. © Authors 2012.

关键词： Electric power utilization

来源：评论

学校读者我要写书评

暂无评论

Computational study of separation and capture of micro-particles in microfluidic devices

Computational study of separation and capture of micro-parti...

引用

International Conference on Advanced Materials for Science and Engineering (ICAMSE)

作者： Heng-Chuan Kan Jian-Ming Lu National Applied Research Laboratories National Center for High-Performance Computing Tainan City Taiwan R.O.C

ISBN: (纸本)9781509038701

Separation and concentration of blood cells are vital steps in biological sample preparation for the design and operation of microfluidic biochips. There have been significant efforts toward developing approaches for effective separation and capture of blood cells in microfluidic devices recently. The present paper is aimed to evaluate separation and capture of micro-particles in passive microfluidic channels using numerical simulations. Two passive layouts of microfluidic channels are employed in this study: imbalance-bifurcation and filter-trench. results reveal that (i) infusion of different micro-particle sizes results in reduction of separation efficiency; (ii) it requires high splitting ratio of bifurcation branch to main channel to achieve high separation rate; (iii) the layout of two consecutive filter-trench microfluidic channel increases the capture rate of micro-particles and achieves better concentration efficiency.

关键词： Microfluidics Blood Bifurcation Cells (biology) Conferences Layout Computational modeling

来源：评论

学校读者我要写书评

暂无评论

Topology-Aware Process Mapping on Clusters Featuring NUMA and Hierarchical Network

Topology-Aware Process Mapping on Clusters Featuring NUMA an...

引用

International Symposium on Parallel and distributed computing

作者： dongyang Li Yunlan Wang Wei Zhu School of Computer Science & Center for High Performance Computing Northwestern Polytechnical University Xi'an Shaanxi P.R.China

ISBN: (纸本)9781479907342

Topology-aware process mapping plays a pivotal role in improving the performance of parallel programs on modern large-scale cluster computing systems featuring multicore architecture and hierarchical network. This paper firstly introduces a method to obtain application communication topology and hardware topology. Then a topology-aware process mapping algorithm named Isomorphic Tree Mapping is proposed for computing systems with hierarchical networks and NUMA effect. This algorithm partitions the sets of processes based on process proximity (or affinity) which is defined considering both the communication volume and frequency. during the partition of processes, a tree-like structure is built which is isomorphic to the physical topology tree. Then a mapping between any two correspondent nodes on the two isomorphic trees is performed. Experimental evaluations on a multicore cluster show that the process mapping method presented in this paper can effectively improve the communication efficiency of MPI application and cut down up to 10% of the whole execution time in most cases.

关键词： Network topology Topology Hardware Bandwidth Sockets Multicore processing

来源：评论

学校读者我要写书评

暂无评论

The impact of assembly history on the X-ray detectability of halos From galaxy groups to galaxy clusters

arXiv

引用

arXiv 2025年

作者： Marini, I. Popesso, P. dolag, K. Biffi, V. Vladutescu-Zopp, S. Castro, T. Toptun, V. de Isídio, N. dev, A. Mazengo, d. Comparat, J. Gouin, C. Malavasi, N. Merloni, A. Mroczkowski, T. Ponti, G. Shreeram, S. Zhang, Y. European Southern Observatory Karl Schwarzschildstrasse 2 Garching bei München85748 Germany Excellence Cluster ORIGINS Boltzmannstr. 2 Garching bei MünchenD-85748 Germany Universitäts-Sternwarte Fakultät für Physik Ludwig-Maximilians-Universität München Scheinerstr.1 München81679 Germany Max-Planck-Institut für Astrophysik Karl-Schwarzschildstr. 1 Garching bei München85741 Germany INAF Osservatorio Astronomico di Trieste Via Tiepolo 11 Trieste34143 Italy IFPU Institute for Fundamental Physics of the Universe Via Beirut 2 TriesteI-34014 Italy INFN Sezione di Trieste Via Valerio 2 TS Trieste34127 Italy ICSC - Centro Nazionale di Ricerca in High Performance Computing Big Data e Quantum Computing Via Magnanelli 2 Bologna Italy Sorbonne Université UMR7095 Institut d’Astrophysique de Paris 98 bis Boulevard Arago ParisF-75014 France Université Paris-Saclay CNRS Institut d’Astrophysique Spatiale Orsay91405 France International Centre for Radio Astronomy Research University of Western Australia M468 35 Stirling Highway PerthWA6009 Australia Department of Physics College of Natural and Mathematical Sciences University of Dodoma P.O. Box 338 Dodoma Tanzania United Republic of Max Planck Institute for Extraterrestrial Physics Giessenbachstrasse 1 Garching85748 Germany INAF Osservatorio Astronomico di Brera Via E. Bianchi 46 LC Merate23807 Italy DiSAT Università degli Studi dell’Insubria via Valleggio 11 ComoI-22100 Italy

Galaxy groups represent a significant fraction of the halo population, playing a crucial role in galaxy formation and evolution. However, their detection in X-rays remains challenging, raising questions about the physical mechanisms driving their detectability in current surveys. Using the Magneticum simulations, we construct a mock X-ray lightcone of the local Universe (z 500 = 1013 M☉ can have hot gas fractions in the range fgas = 0.02 − 0.06 and a central SMBH with a median mass of MBH = 109 M☉ and a scatter of 0.5 dex) compatible with the most recent measurements of the baryonic fraction. These findings will aid the interpretation of future X-ray surveys, demonstrating the power of simulation-based inference. Copyright © 2025, The Authors. All rights reserved.

关键词： Black holes

来源：评论

学校读者我要写书评

暂无评论

Using dASPK on the TMC CM5. Experiences with two programming models

Using DASPK on the TMC CM5. Experiences with two programming...

引用

1993 Scalable Parallel Libraries Conference, SPLC 1993

作者： Maier, r.S. Petzold, L.r. rath, W. Minnesota Supercomputer Center Inc. Army High Performance Computing Research Center 1100 Washington Avenue South MinneapolisMN55415 United States Department of Computer Science University of Minnesota MinneapolisMN55455 United States Fachbereich Mathematik TU Chemnitz-Zwickau ChemnitzD-09009 Germany

ISBN: (纸本)0818649801

dASPK solves large-scale systems of differential-algebraic equations. We have developed two parallel versions of dASPK. They are dASPKF90, a Fortran 90 data parallel implementation for the cmf compiler, and dASPKMP, a message-passing implementation written in Fortran 77 with extended BLAS. We describe the two codes, demonstrate the application of the codes on true large-scale examples, gave some timing results, and draw some conclusions based on our experiences. © 1994 IEEE.

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：