检索结果-内蒙古大学图书馆

A reconfigurable routing method for fault-tolerant mesh-based network on chip

Journal of Information and Computational Science 2013年第1期10卷 157-165页

作者： Liu, Yanhua Ruan, Ying Lai, Zongsheng Sun, Ling Institute of Microelectronics Circuit and System East China Normal University Shanghai 200241 China Jiangsu Provincial Key Lab. of ASIC Design Nantong University Nantong 226019 China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China

As the semiconductor industry advances to nano-technology points, Network on Chip (NoC) components are becoming vulnerable to errors during the system operation. Consequently, fault-tolerant techniques for NoC are needed to achieve high reliability. In this paper, we present a fault-tolerant routing method to tolerate faulty routers in a two-dimension (2D) mesh-based NoC. With this method, the system can have a multi-router fault-tolerant capability and the routing path can be dynamically reconfigured to support different situations caused by faulty routers in NoC. Fault-tolerance capability and system performance of the proposed method are evaluated through analysis and simulations. Theoretical analysis and simulation results show that the proposed method can enhance the reliability of system and offer considerable fault-tolerance capability. © 2013 by Binary Information Press.

关键词： Network-on-chip

来源：评论

学校读者我要写书评

暂无评论

Supervisor synthesis and performance improvement for automated manufacturing systems by using Petri nets

Supervisor synthesis and performance improvement for automat...

引用

2013 IEEE International Conference on Automation Science and Engineering, CASE 2013

作者： Hu, Hesuan Zhou, Mengchu Liu, Yang School of Electro-mechanical Engineering Xidian University Xi'an Shaanxi 710071 China School of Computer Engineering Nanyang Technological University Singapore 639798 Singapore Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai 201804 China Department of Electrical and Computer Engineering New Jersey Institute of Technology Newark NJ 07102 United States

ISBN: (纸本)9781479915156

Behavior management and performance evaluation are two significant issues in the design and operation of automated manufacturing systems (AMS). The former constrains their dynamics so as to avoid improper behavior and eliminate downtime, while the latter aims to evaluate cycle time. Although abundant research has been conducted on each issue independently, researchers rarely address both in a correlative way. Without accurate and prompt guidance to route parts, it is difficult for an AMS to manage the production process in an optimal manner. In the paradigm of Petri nets, we show that it is crucial to optimize certain local priorities for the sake of cycle time minimization or throughput maximization. Without making radical changes to the system structure, our method can well design some control parameters in order to meet system performance requirement. An algorithm of polynomial complexity is presented to derive supervisory controllers, if existing, to improve system performance. © 2013 IEEE.

关键词： Petri nets

来源：评论

学校读者我要写书评

暂无评论

Maximally permissive distributed control of large scale automated manufacturing systems modeled with Petri nets

Maximally permissive distributed control of large scale auto...

引用

2013 IEEE International Conference on Automation Science and Engineering, CASE 2013

ISBN: (纸本)9781479915156

Ensuring nonblockingness remains challenging for automated manufacturing systems (AMS) owing to their discrete event dynamics. Both scalability and maximal permissiveness are essential for the synthesis and implementation of their centralized supervisors. Inspired by the divide and conquer philosophy, this work proposes a partition methodology and distributed control technique for large scale AMS. They are represented as interconnected and overlapping subsystems sharing some common components in terms of buffers. For each subsystem, a local supervisor is designed based on its local behavior and neighboring information only. Generalizing the existing results, we develop a condition under which the control law via decomposition promises the maximal permissiveness. Buffer capacities are well designed for the sake of their decomposition into multiple overlapping subsystems. Theoretical results are developed to characterize the behavior compatibility among local controllers. An experimental study illustrates the effectiveness of the proposed method. © 2013 IEEE.

关键词： Supervisory personnel

来源：评论

学校读者我要写书评

暂无评论

Web-Age Information Management 1

引用

丛书名： Lecture Notes in computer Science

1000年

作者： Yunjun Gao Kyuseok Shim Zhiming Ding Peiquan Jin Zujie Ren Yingyuan Xiao An Liu Shaojie Qiao

ISBN: (数字)9783642395277

ISBN: (纸本)9783642395260

This book constitutes the refereed proceedings of six workshops of the 14th International Conference on Web-Age Information Management, WAIM 2013, held in Beidaihe, China, June 2013. The 37 revised full papers are organized in topical sections on the six following workshops: The International Workshop on Big Data Management on Emerging Hardware (HardBD 2013), the Second International Workshop on Massive Data Storage and Processing (MDSP 2013), the First International Workshop on Emergency Management in Big Data Age (BigEM 2013), the International Workshop on Trajectory Mining in Social Networks (TMSN 2013), the First International Workshop on Location-based Query Processing in Mobile Environments (LQPM 2013), and the First International Workshop on Big Data Management and Service (BDMS 2013).

关键词： Information Storage and Retrieval Information systems Applications (incl. Internet) Data Mining and Knowledge Discovery Database Management computer Communication Networks Algorithm Analysis and Problem Complexity

来源：评论

学校读者我要写书评

暂无评论

Automated web service composition supporting conditional branch structures

引用

Enterprise Information systems 2014年第1期8卷 121-146页

作者： Wang, Pengwei Ding, Zhijun zhijun_ding@*** Jiang, Changjun Zhou, Mengchu Department of Computer Science and Technology Tongji University Shanghai 201804 China Key Laboratory of “Embedded System and Service Computing” Ministry of Education Shanghai 201804 China Department of Electrical and Computer Engineering New Jersey Institute of Technology Newark NJ 07102 USA

The creation of value-added services by automatic composition of existing ones is gaining a significant momentum as the potential silver bullet in service-oriented architecture. However, service composition faces two aspects of difficulties. First, users' needs present such characteristics as diversity, uncertainty and personalisation; second, the existing services run in a real-world environment that is highly complex and dynamically changing. These difficulties may cause the emergence of nondeterministic choices in the process of service composition, which has gone beyond what the existing automated service composition techniques can handle. According to most of the existing methods, the process model of composite service includes sequence constructs only. This article presents a method to introduce conditional branch structures into the process model of composite service when needed, in order to satisfy users' diverse and personalised needs and adapt to the dynamic changes of real-world environment. UML activity diagrams are used to represent dependencies in composite service. Two types of user preferences are considered in this article, which have been ignored by the previous work and a simple programming language style expression is adopted to describe them. Two different algorithms are presented to deal with different situations. A real-life case is provided to illustrate the proposed concepts and methods. [ABSTRACT FROM AUTHOR]

关键词： conditional branch structure control construct service composition user preference web service

来源：评论

学校读者我要写书评

暂无评论

MALK: A MapReduce framework for high-efficiently processing large amount of keys

引用

Jisuanji Yanjiu yu Fazhan/computer Research and Development 2014年第12期51卷 2711-2723页

作者： Zheng, Yasong Wang, Da Ye, Xiaochun Cui, Huimin Xu, Yuanchao Fan, Dongrui State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China College of Information Engineering Capital Normal University Beijing100048 China

The overhead of memory allocation is one of the major bottlenecks for shared-memory MapReduce, especially for the applications that have large amount of keys. In order to solve this problem, this paper presents a less memory consumption MapReduce, namely MALK, which can high-efficiently process applications with a large number of keys. Firstly, MALK succeeds in avoiding the constant allocations of massive small memory blocks by managing the discrete keys using contiguous area of storage. Secondly, MALK pipelines the process of Map-tasks and Reduce-tasks to decrease the active data in the system at the same time, and proposes a reusable mechanism of Hash table to reuse the memory space so as to avoid the memory reallocation of Hash table. What is more, MALK determines the suitable number of Reduce tasks, by evaluating the effect of task quantity and granularity on performance, to get optimal performance. The experiments show that, compared with Phoenix++, MALK achieves up to 3.8X higher speedup (average of 2.8X), and saves up to 95.2% memory in Map phase and 87.8% memory in Reduce phase. In addition, MALK reduces 30% waiting time with better load balance in Reduce phase, and cuts down more than 35% cache miss rate on average. ©, 2014, Science Press. All right reserved.

关键词： MapReduce

来源：评论

学校读者我要写书评

暂无评论

Lightdet: A Lightweight and Accurate Object Detection Network

Lightdet: A Lightweight and Accurate Object Detection Networ...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Qiankun Tang Jie Li Zhiping Shi Yu Hu Research Center for Intelligent Computing Systems State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing Advanced Innovation Center for Imaging Theory and Technology Capital Normal University

ISBN: (数字)9781509066315

ISBN: (纸本)9781509066322

The extensive computational burden limits the usage of accurate but complex object detectors in resource-bounded scenarios. In this paper, we present a lightweight object detector, named LightDet, to address this dilemma. We design a lightweight backbone that is able to capture rich low-level features by the proposed Detail-Preserving Module. To effectively aggregate bottom and top-down features, we introduce an efficient Feature-Preserving and Refinement Module. A lightweight prediction head is employed to further reduce the entire network complexity. Experimental results show that our LightDet achieves 75.5% mAP on PASCAL VOC 2007 at the speed of 250 FPS and 24.0% mAP on MS COCO dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Energy-performance analysis and optimization for networks-on-chip

引用

Jisuanji Xuebao/Chinese Journal of computers 2013年第5期36卷 988-1003页

作者： Zhang, Shuai Song, Feng-Long Wang, Dong Liu, Zhi-Yong Fan, Dong-Rui State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China University of Chinese Academy of Sciences Beijing 100039 China State Grid Information and Telecommunication Ltd. Beijing 100761 China

This paper discusses energy-performance trade-off of networks-on-chip with real parallel applications. First, we develop an accurate energy-performance analytical model that, and conduct and analyze the impacts of both frequency-independent and frequency-dependent factors in energy and performance model. Secondly, we put together the communication overhead, memory access overhead, frequency scaling, core count scaling to quantify the performance and energy consumed by NoCs in the energy-performance model. We propose a new energy-performance optimization method, by choosing a pair of frequency and core count to get optimal energy or performance. Finally, we implement eight PARSEC parallel applications to evaluate our model and optimization method. The experiment results confirm that our model predicts NoCs energy and performance behavior well, and we exactly select correct frequency level and core count for minimum energy or maximum performance with a given constraint for most parallel applications. We also present how to establish an NoCs' energy-performance model of a real parallel application in this paper.

关键词： Network-on-chip

来源：评论

学校读者我要写书评

暂无评论

ENAS4D: Efficient multi-stage CNN architecture search for dynamic inference

arXiv

引用

arXiv 2020年

作者： Yuan, Zhihang Liu, Xin Wu, Bingzhe Sun, Guangyu Peking University Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences China

Dynamic inference is a feasible way to reduce the computational cost of convolutional neural network(CNN), which can dynamically adjust the computation for each input sample. One of the ways to achieve dynamic inference is to use multistage neural network, which contains a sub-network with prediction layer at each stage. The inference of a input sample can exit from early stage if the prediction of the stage is confident enough. However, design a multi-stage CNN architecture is a non-trivial task. In this paper, we introduce a general framework, ENAS4D, which can efficiently search for optimal multi-stage CNN architecture for dynamic inference in a well-designed search space. Firstly, we propose a method to construct the search space with multi-stage convolution. The search space include different numbers of layers, different kernel sizes and different numbers of channels for each stage and the resolution of input samples. Then, we train a once-for-all network that supports to sample diverse multi-stage CNN architecture. A specialized multi-stage network can be obtained from the once-for-all network without additional training. Finally, we devise a method to efficiently search for the optimal multi-stage network that trades the accuracy off the computational cost taking the advantage of once-for-all network. The experiments on the ImageNet classification task demonstrate that the multi-stage CNNs searched by ENAS4D consistently outperform the state-of-the-art method for dyanmic inference. In particular, the network achieves 74.4% ImageNet top-1 accuracy under 185M average MACs. Copyright © 2020, The Authors. All rights reserved.

关键词： Memory architecture

来源：评论

学校读者我要写书评

暂无评论

Rubik: A hierarchical architecture for efficient graph learning

arXiv

引用

arXiv 2020年

作者： Chen, Xiaobing Wang, Yuke Xie, Xinfeng Hu, Xing Basak, Abanti Liang, Ling Yan, Mingyu Deng, Lei Ding, Yufei Du, Zidong Chen, Yunji Xie, Yuan State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Beijing100190 China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China Department of Computer Science University of California Santa Barbara United States Department of Electrical and Computer Engineering University of California Santa Barbara United States

Graph convolutional network (GCN) emerges as a promising direction to learn the inductive representation in graph data commonly used in widespread applications, such as E-commerce, social networks, and knowledge graphs. However, learning from graphs is non-trivial because of its mixed computation model involving both graph analytics and neural network computing. To this end, we decompose the GCN learning into two hierarchical paradigms: graph-level and node-level computing. Such a hierarchical paradigm facilitates the software and hardware accelerations for GCN learning. We propose a lightweight graph reordering methodology, incorporated with a GCN accelerator architecture that equips a customized cache design to fully utilize the graph-level data reuse. We also propose a mapping methodology aware of data reuse and task-level parallelism to handle various graphs inputs effectively. Results show that Rubik accelerator design improves energy efficiency by 26.3x to 1375.2x than GPU platforms across different datasets and GCN models. Copyright © 2020, The Authors. All rights reserved.

关键词： Graph neural networks

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：