检索结果-内蒙古大学图书馆

A recursive-adjustment co-allocation scheme in data grid environments 1

6th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP

作者： Yang, Chao-Tung Yang, I.-Hsien Li, Kuan-Ching Hsu, Ching-Hsien High Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Parallel and Distributed Processing Center Department of Computer Science and Information Management Providence University Taichung 43301 Taiwan Department of Computer Science and Information Engineering Chung Hua University Hsinchu 300 Taiwan

ISBN: (数字)9783540320715

ISBN: (纸本)3540292357

The co-allocation architecture was developed in order to enable parallel downloads of datasets from multiple servers. Several co-allocation strategies have been coupled and used to exploit rate differences among various client-server links and to address dynamic rate fluctuations by dividing files into multiple blocks of equal sizes. However, a major obstacle, the idle time of faster servers having to wait for the slowest server to deliver the final block, makes it important to reduce differences in finish time among replica servers. In this paper, we propose a dynamic co-allocation scheme, namely Recursive-Adjustment Co-Allocation scheme, to improve the performance of data transfer in Data Grids. Our approach reduces the idle time spent waiting for the slowest server and decreases data transfer completion time. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Neural generative question answering 25

Neural generative question answering

引用

25th International Joint Conference on Artificial Intelligence, IJCAI 2016

作者： Yin, Jun Xin, Jiang Lu, Zhengdong Shang, Lifeng Li, Hang Li, Xiaoming School of Electronic Engineering and Computer Science Peking University China Noah's Ark Lab Huawei Technologies China Collaborative Innovation Center of High Performance Computing NUDT Changsha China

This paper presents an end-to-end neural network model, named Neural Generative Question Answering (GENQA), that can generate answers to simple factoid questions, based on the facts in a knowledge-base. More specifically, the model is built on the encoder-decoder framework for sequence-to-sequence learning, while equipped with the ability to enquire the knowledge-base, and is trained on a corpus of question-answer pairs, with their associated triples in the knowledge-base. Empirical study shows the proposed model can effectively deal with the variations of questions and answers, and generate right and natural answers by referring to the facts in the knowledge-base. The experiment on question answering demonstrates that the proposed model can outperform an embedding-based QA model as well as a neural dialogue model trained on the same data.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

A chronological history-based execution time estimation model for embarrassingly parallel applications on grids

A chronological history-based execution time estimation mode...

引用

3rd International Symposium on Parallel and Distributed Processing and Applications, ISPA 2005

作者： Yang, Chao-Tung Shih, Po-Chi Lin, Cheng-Fang Hsu, Ching-Hsien Li, Kuan-Ching High Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Department of Computer Science and Information Engineering Chung Hua University Hsinchu 300 Taiwan Parallel and Distributed Processing Center Department of Computer Science and Information Management Providence University Taichung 43301 Taiwan

ISBN: (纸本)3540297693

In order to identify and schedule jobs that are suitable for determined resources, an execution time estimation model is required. In this paper, it is described a Chronological history-based execution time estimation model to predict current execution time, according to the previous execution results. We built a heterogeneous computational Grid environment using Globus Toolkit, and our research is focused in Grid computing environments and to execute parallel jobs on multiple resources by measuring its accuracy. The experimental results shown that our model can accurately predict the execution time of embarrassingly parallel applications. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Distributed computer systems

来源：评论

学校读者我要写书评

暂无评论

On utilization of the grid computing technology for video conversion and 3D rendering

On utilization of the grid computing technology for video co...

引用

3rd International Symposium on Parallel and Distributed Processing and Applications, ISPA 2005

作者： Yang, Chao-Tung Lai, Chuan-Lin Li, Kuan-Ching Hsu, Ching-Hsien Chu, William C. High Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Parallel and Distributed Processing Center Department of Computer Science and Information Management Providence University Taichung 43301 Taiwan Department of Computer Science and Information Engineering Chung Hua University Hsinchu 300 Taiwan

ISBN: (纸本)3540297693

In this paper, we investigate the recent popular computing technique called Grid computing, and use video conversion and 3D rendering applications to demonstrate this technology's effectiveness and high performance. We also report on developing a resource broker called Phantom that runs on our grid computing testbed and whose main function is querying nodes in grid computing environments and showing their system information to aid in selecting the best nodes for job assignments to have the jobs executed in the least amount of time. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Computability and decidability

来源：评论

学校读者我要写书评

暂无评论

Scalability analysis and evaluation of divisible load scheduling 43

Scalability analysis and evaluation of divisible load schedu...

引用

43rd International Conference on Parallel Processing Workshops, ICPPW 2014

作者： Balasubramaniam, Mahadevan Banicescu, Ioana Ciorba, Florina M. Department of Computer Science and Engineering Mississippi State University MS39762 United States Center for Information Services and High Performance Computing Technische Universität Dresden01062 Germany

ISBN: (纸本)9781479956159

In this work, we address the problem of the scalability of divisible load scheduling of data parallel workloads (also called arbitrarily divisible workloads) on high performance parallel and distributed computing systems. Divisible load theory offers a linear, deterministic, and tractable model for scheduling arbitrarily divisible workloads, often encountered in scientific applications. With the continuous increasing problem and the system sizes, it is imperative that application scheduling algorithms scale well to leverage the processing capabilities of the high performance computing systems. We conduct an analytical evaluation as well as a simulation-based study of the scalability of divisible load scheduling algorithms, called DLT algorithms, when applied for scheduling two NAS parallel benchmarks, namely, the embarrassingly parallel (EP) and the integer sort (IS) benchmarks onto a target system modeled as a 3-d torus topology. The EP benchmark is computationally intensive and the IS benchmark is communication intensive. Two questions related to the scalability study are addressed, namely, the fastest time to solve the problem and the condition for cost-optimality. A simulation-based study was conducted to support and verify the solutions obtained via the analytical model. © 2014 IEEE.

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

Visuel: A novel performance monitoring and analysis toolkit for cluster and grid environments

引用

6th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP

作者： Li, Kuan-Ching Cheng, Hsiang-Yao Yang, Chao-Tung Hsu, Ching-Hsien Wang, Hsiao-Hsi Hsu, Chia-Wen Hung, Sheng-Shiang Chang, Chia-Fu Liu, Chun-Chieh Pan, Yu-Hwa Parallel and Distributed Processing Center Department of Computer Science and Information Management Providence University Taichung 43301 Taiwan High Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Department of Computer Science and Information Engineering Chung Hua University Hsinchu 300 Taiwan

ISBN: (纸本)3540292357

The computing power provided by high performance low-cost PC-based Cluster and Grid platforms are attractive, and they are equal or superior to supercomputers and mainframes widely available. In this research paper, we present the design rationale and implementation of Visuel, a toolkit for performance measurement and analysis of MPI parallel programs and real time resources monitoring in cluster and grid computing environments. The proposed toolkit is web-based interface to show performance activities of all computing nodes involved in the execution of a MPI parallel program, such as CPU and memory usage levels of each computing node, and monitors all computing nodes of a computing platform by displaying real time performance data. In addition, this toolkit is able to display comparative performance data charts of multiple executions of MPI parallel application under investigation, which facilitates the "what-if" analysis. The usage of this toolkit shows that it outperforms in easing the process of investigation of parallel applications. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Distributed computer systems

来源：评论

学校读者我要写书评

暂无评论

Efficient scheduling of recursive control flow on GPUs 13

Efficient scheduling of recursive control flow on GPUs

引用

27th ACM International Conference on Supercomputing, ICS 2013

作者： Huo, Xin Krishnamoorthy, Sriram Agrawal, Gagan Department of Computer Science and Engineering Ohio State University Columbus OH United States High Performance Computing Group Pacific Northwest National Laboratory Richland WA United States

ISBN: (纸本)9781450321303

Graphics processing units (GPUs) have rapidly emerged as a very significant player in high performance computing. Single instruction multiple thread (SIMT) pipelines are typically used in GPUs to exploit parallelism and maximize performance. Although support for unstructured control flow has been included in GPUs, efficiently managing thread divergence for arbitrary parallel programs remains a critical challenge. In this paper, we focus on the problem of supporting recursion in modern GPUs. We design and comparatively evaluate various algorithms to manage thread divergence encountered in recursive programs. The results improve upon traditional post-dominator based reconvergence mechanisms designed to handle thread divergence due to control flow within a procedure. © 2013 ACM.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Gradient Amplification: An Efficient Way to Train Deep Neural Networks

引用

Big Data Mining and Analytics 2020年第3期3卷 196-207页

作者： Sunitha Basodi Chunyan Ji Haiping Zhang Yi Pan Department of Computer Science Georgia State UniversityAtlantaGA 30302USA Center for High Performance Computing Joint Engineering Research Center for Health Big Data Intelligent Analysis TechnologyShenzhen Institutes of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China

Improving performance of deep learning models and reducing their training times are ongoing challenges in deep neural *** are several approaches proposed to address these challenges,one of which is to increase the depth of the neural *** deeper networks not only increase training times,but also suffer from vanishing gradients problem while *** this work,we propose gradient amplification approach for training deep learning models to prevent vanishing gradients and also develop a training strategy to enable or disable gradient amplification method across several epochs with different learning *** perform experiments on VGG-19 and Resnet models(Resnet-18 and Resnet-34),and study the impact of amplification parameters on these models in *** proposed approach improves performance of these deep learning models even at higher learning rates,thereby allowing these models to achieve higher performance with reduced training time.

关键词： deep learning gradient amplification learning rate backpropagation vanishing gradients

来源：评论

学校读者我要写书评

暂无评论

Grid computing in Taiwan

Grid computing in Taiwan

引用

Proceedings - 10th IEEE International Workshop on Future Trends of Distributed computing Systems

作者： Yang, Chao-Tung Chu, William C. High-Performance Computing Lab. Department of Computer Science Tunghai University Taichung 407 Taiwan Software Eng. and Technology Center Department of Computer Science Tunghai University Taichung 407 Taiwan

ISBN: (纸本)0769521185

Grid computing focuses on aggregating resources (e.g., processor cycles, disk storage and contents) from a large-scale computing environment. It intends to deliver high-performance distributed platforms for computation- and/or data-intensive applications. In this paper, we will study the enabling techniques for grid computing for high-performance computing. Our goals are to (1) understand the design of these key components, (2) set up a grid computing platform, (3) learn how to create grid-enabled high-performance applications, and (4) share experiences on constructing such platforms and applications from the status of Taiwan.

关键词： Computational methods

来源：评论

学校读者我要写书评

暂无评论

An efficient network information model using NWS for grid computing environments

引用

4th International Conference on Grid and Cooperative computing - GCC 2005

作者： Yang, Chao-Tung Shih, Po-Chi Chen, Sung-Yi Shih, Wen-Chung High-Performance Computing Laboratory Department of Computer Science and Information Engineering Tunghai University Taichung 40704 Taiwan Departments Computer Science National Tsing Hua University Hsinchu 30013 Taiwan Department of Computer and Information Science National Chiao Tung University Hsinchu 300 Taiwan

ISBN: (纸本)3540305106

Grid computing technologies enable large-scale aggregation and sharing of resources via wide-area networks focused on sharing computational, data, and other resources to form general-purpose services for users. In this paper, we address network information gathering and focus on providing approximate measurement models for network-related information using Network Weather Service (NWS) for future scheduling and benchmarking. We propose a network measurement model for gathering network-related information including bandwidth, latency, forecasting, error rates, etc., without generating excessive system overhead. We consider inaccuracies in real-world network values in generating approximation values for future use. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Virtual reality

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：