检索结果-内蒙古大学图书馆

ICETC 2010 - 2010 2nd International Conference on Education Technology and Computer 2010年 4卷 V4311-V4315页

作者： Li, Huiba Peng, Yuxing Lu, Xicheng National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha 410073 China

ISBN: (纸本)9781424463688

Event-driven programming has been a relatively hot topic in distributed systems development. Having worked on these systems for years, we now believe that it is not the best choice. Besides the well-known "stack ripping" problem, we argue that it greatly influences the composability of software modules. Preemptive threads are also short of composability because of data-races and locks. Lacking of composability can result in systems with little vitality. Cooperative threading (or coroutine), on the contrary, is almost free of this problem, so we advocate it as the primary concurrency model for most distributed systems. © 2010 IEEE.

关键词： distributed computer systems

来源：评论

学校读者我要写书评

暂无评论

A blind motion deblurring for daytime and nighttime images

A blind motion deblurring for daytime and nighttime images

引用

International Conference on Wavelet Active Media Technology and Information processing (ICWAMTIP)

作者： Fei Zhang Yong Dou Xin Niu National Laboratory for Parallel and Distributed Processing National University of Defense Technology

ISBN: (纸本)9781479972098

This paper presents an automatic deblurring approach for motion blur images. The approach explores the prior of the intensity and gradient to estimate the motion blur kernel from single blurred image. In this way, motion blur kernel could be well estimated not only on daytime images, but also on nighttime images. Efficient optimization method was given for the prior-based approach. Besides, a cost-effective method was also proposed to remove the noise of the kernel in the iterative kernel estimation. The proposed kernel enhancement approach could also improve the performance of other blind deblurring algorithm. Experimental results showed that the performance of the proposed algorithm was superior to many state-of-the-art blind deblurring algorithms, especially for nighttime images with saturated regions.

关键词： Kernel Estimation Deconvolution Noise Image edge detection Image restoration Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

Pruning false positives of static data-race detection via thread specialization

Pruning false positives of static data-race detection via th...

引用

10th International Symposium on Advanced parallel processing Technologies, APPT 2013

作者： Chen, Chen Lu, Kai Wang, Xiaoping Zhou, Xu Fang, Li National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha China College of Computer National University of Defense Technology Changsha China Information Center National University of Defense Technology Changsha China

ISBN: (纸本)9783642452925

Static data-race detection is a powerful tool by providing clues for dynamic approaches to only instrument certain memory accesses. However, static data-race analysis suffers from high false positive rate. A key reason is that static analysis overestimates the set of shared objects a thread can access. We propose thread specialization to distinguish threads statically. By fixing the number of threads as well as the ID assigned to each thread, a program can be transformed to a simplified version. Static data-race analysis on this simplified program can infer the range of addresses accessed by each thread more accurately. Our approach prunes false positives by an average of 89.2% and reduces dynamic instrumentation by an average of 63.4% in seven benchmarks. © 2013 Springer-Verlag.

关键词： Static analysis

来源：评论

学校读者我要写书评

暂无评论

Simulation research of communication networks based on prim algorithm

Simulation research of communication networks based on prim ...

引用

International Forum on Information Technology and Applications

作者： Yang, Chenghui Ren, ChuanJun College of Electrical Engineering Northwest University for Nationalities Lanzhou 730030 China National Laboratory for Parallel and Distributed Processing College of Computer Science National University of Defense Technology Changsha 410073 China

ISBN: (纸本)9780769541150

In order to improve the efficiency of the communication networks, we used the Kruskal algorithm and the Prim algorithm through algorithm comparison and analysis methods of data structure. A dynamic framework for the communication network installed system is built. Moreover, according to the actual framework of the communication network specific issues, the module chooses the Minimum Cost Spanning Tree Prim algorithm ultimately. The assumptions process and outcomes simulation have proper analysis and certification by C language. Experimental results show that the Minimum Cost Spanning Tree of Prim algorithm on the installed system for communication networks dynamic planning process have the advantage of speed and effectively reduce the waste of resources, which not only can ensure efficiency but also can effectively improve communication networks installed cost. © 2010 IEEE.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

A clustering model for memory resource sharing in large scale distributed system

A clustering model for memory resource sharing in large scal...

引用

International Conference on parallel and distributed Systems (ICPADS)

作者： Rui Chu Nong Xiao Xicheng Lu National Laboratory for Parallel and Distributed Processing ChangSha Hunan China

As an application of large scale distributed network computing system, RAM Grid tries to solve the problem of memory resource sharing and utilization. Due to the special properties of memory, traditional resource information management approaches cannot be adapted easily. This paper proposes a clustering based resource aggregating scheme under the background of RAM Grid, which can reduce the scale of resource information management efficiently. With analogy to the force field and potential energy theory in physics, the basic model, the force field-potential energy model, and the corresponding distributed algorithms are proposed, respectively. The model and algorithms are also evaluated by real network topologies based simulation.

关键词： Resource management Large-scale systems Random access memory Read-write memory Information management Computer networks distributed computing Grid computing Memory management Potential energy

来源：评论

学校读者我要写书评

暂无评论

64-bit floating-point FPGA matrix multiplication 05

64-bit floating-point FPGA matrix multiplication

引用

ACM/SIGDA Thirteenth ACM International Symposium on Field Programmable Gate Arrays - FPGA 2005

作者： Dou, Yong Vassiliadis, S. Kuzmanov, G.K. Gaydadjiev, G.N. National Laboratory for Parallel and Distributed Processing Changsha 410073 China Computer Engineering EEMCS TU Delft P.O. Box 5031 2600 GA Delft Netherlands

ISBN: (纸本)9781595930293

We introduce a 64-bit ANSI/IEEE Std 754-1985 floating point design of a hardware matrix multiplier optimized for FPGA implementations. A general block matrix multiplication algorithm, applicable for an arbitrary matrix size is proposed. The algorithm potentially enables optimum performance by exploiting the data locality and reusability incurred by the general matrix multiplication scheme and considering the limitations of the I/O bandwidth and the local storage volume. We implement a scalable linear array of processing elements (PE) supporting the proposed algorithm in the Xilinx Virtex II Pro technology. Synthesis results confirm a superior performance-area ratio compared to related recent works. Assuming the same FPGA chip, the same amount of local memory, and the same I/O bandwidth, our design outperforms related proposals by at least 1.7X and up to 18X consuming the least reconfigurable resources. A total of 39 PEs can be integrated into the xc2vp125-7 FPGA, reaching performance of, e.g., 15.6 GFLOPS with 1600 KB local memory and 400 MB/s external memory bandwidth. Copyright 2005 ACM.

关键词： Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

A Data-Centric Approach for Efficient and Scalable CFD Implementation on Multi-GPUs Clusters 24th

A Data-Centric Approach for Efficient and Scalable CFD Imple...

引用

24th International Conference on parallel and distributed Computing, Applications and Technologies, PDCAT 2023

作者： Li, Ruitian Deng, Liang Dai, Zhe Zhang, Jian Liu, Jie Liu, Gang China Aerodynamic Research and Development Center Computational Aerodynamic Institute Mianyang China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha China

ISBN: (纸本)9789819982103

Scalability is a crucial factor determining the performance of massive heterogeneous parallel CFD applications on the multi-GPUs platforms, particularly after the single-GPU implementations have achieved optimal performance through numerous optimizations. A novel Data-Centric hybrid MPI-CUDA CFD model is proposed in this paper to enable efficient scalability of CFD applications on large-scale heterogeneous platforms. Based on the Data-Centric approach, Minimum-cost MPI transfer strategy and the code refactoring technique are realized for a better balance between data transfer and floating-point computation performance, which could significantly improve the scalability and reduce the time-to-solution. Subsequently, those approaches are integrated into the industrial unstructured CFD software, FlowStar, to evaluate their effectiveness. Numerical results demonstrate that Minimum-cost MPI strategy achieves more than 2.0 times performance improvement compared to the traditional Model-Centric implementation, and the code refactoring technique boosts performance by 40% to 50% over the minimum-cost MPI version. Moreover, the Data-Centric implementation on 64 A100 GPUs platform show a speedup ratio of over 120 when compared to the original MPI implementation with 64 ranks. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

Crowd intelligence in AI 2.0 era

引用

Frontiers of Information Technology & Electronic Engineering 2017年第1期18卷 15-43页

作者： Wei LI Wen-jun WU Huai-min WANG Xue-qi CHENG Hua-jun CHEN Zhi-hua ZHOU Rong DING State Key Laboratory of Software Development Beihang University National Laboratory for Parallel and Distributed Processing College of ComputerNational University of Defense Technology Institute of Computing Technology Chinese Academy of Sciences College of Computer Science and Technology Zhejiang University National Key Laboratory for Novel Software Technology Nanjing University

The Internet based cyber-physical world has profoundly changed the information environment for the development of artificial intelligence(AI), bringing a new wave of AI research and promoting it into the new era of AI 2.0. As one of the most prominent characteristics of research in AI 2.0 era, crowd intelligence has attracted much attention from both industry and research communities. Specifically, crowd intelligence provides a novel problem-solving paradigm through gathering the intelligence of crowds to address challenges. In particular, due to the rapid development of the sharing economy, crowd intelligence not only becomes a new approach to solving scientific challenges, but has also been integrated into all kinds of application scenarios in daily life, e.g., online-tooffline(O2O) application, real-time traffic monitoring, and logistics management. In this paper, we survey existing studies of crowd intelligence. First, we describe the concept of crowd intelligence, and explain its relationship to the existing related concepts, e.g., crowdsourcing and human computation. Then, we introduce four categories of representative crowd intelligence platforms. We summarize three core research problems and the state-of-the-art techniques of crowd intelligence. Finally, we discuss promising future research directions of crowd intelligence.

关键词： Crowd intelligence Artificial intelligence 2.0 Crowdsourcing Human computation

来源：评论

学校读者我要写书评

暂无评论

Towards an agent oriented programming language with caste and scenario mechanisms

Towards an agent oriented programming language with caste an...

引用

4th International Conference on Autonomous Agents and Multi agent Systems, AAMAS 05

作者： Wang, Ji Shen, Rui Zhu, Hong National Laboratory for Parallel and Distributed Processing Changsha 410073 China Department of Computing Oxford Brookes University Oxford OX33 1HX United Kingdom

The paper presents an agent-oriented programming language SLABSp. It provides caste and scenario mechanisms in a coherent way to support the caste-centric methodology of agent-oriented software development. It uses caste as a modular facility to organize agents into castes and to represent their structure and behavior characteristics. SLABSp also uses scenarios to define agents' behaviors in the context of environment situations. In the paper, the implementation of the language is briefly described. An example of the program is given to illustrate its programming style. Copyright 2005 ACM.

关键词： Computer programming languages

来源：评论

学校读者我要写书评

暂无评论

iRank: Supporting Proximity Ranking for Peer-to-Peer Applications

iRank: Supporting Proximity Ranking for Peer-to-Peer Applica...

引用

International Conference on parallel and distributed Systems (ICPADS)

作者： Yongquan Fu Yijie Wang National Key Laboratory of Parallel and Distributed Processing School of Computer National University of Defense Technology China

Proximity ranking according to end-to-end network distances (e.g., Round-Trip Time, RTT) can reveal detailed proximity information, which is important in network management and performance diagnosis in distributed systems. However, to the best of our knowledge, there has been no similar work on this subject in the P2P computing field. We present a distributed rating method iRank, that enables proximity rankings by providing discrete ratings in a distributed manner. It formulates the proximity ranking as a rating problem that faithfully captures the proximity based on noisy distance measurements scalably and practically. The primary challenge in inferring proximity rankings is enforcing distributed ratings with complex rating policies. Our solution is based on reconstructing ratings by decomposing a centralized rating method Maximum Margin Matrix Factorization (MMMF) into independent sub-problems, that can be efficiently solved in a decentralized manner. By relaxing the dependence on infrastructure nodes that are a single point of failure and limit scalability, iRank can gracefully handle network churns. Through real network latency data sets, we demonstrate that iRank can predict ratings with low distortion, which are smaller than 20 percentage worse than the centralized method, in the context of synthetic complex rating policies.

关键词： Peer to peer computing distributed computing Computer networks Network servers Application software Laboratories distributed processing Concurrent computing Computer network management Conference management

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：