检索结果-内蒙古大学图书馆

作者： Chen, Fuxing Liu, Weiyang Li, Hui Wu, Dongcheng Shenzhen Engineering Lab of Converged Networks Technology Shenzhen Key Lab of Cloud Computing Technology and Application Shenzhen Graduate School Shenzhen 518055 China Graduate School of Information Production and Systems Waseda University Kitakyushu-shi 808- 0135 Japan

The traditional multicast switch fabrics, which were mainly developed from the unicast switch fabrics, currently are not able to achieve high efficiency and flexible large-scale scalability. In the light of lattice theory and multicast concentrator, a novel multistage interconnection multicast switch fabric is proposed in this paper. Comparing to traditional multicast switch fabrics, this multicast switch fabric has the advantages of superior scalability, wire-speed, jitter-free multicast with low delay, and no queuing buffer. This paper thoroughly analyzes the performance of the proposed multicast switch fabric with supporting prioritybased multicast. Simulations on packet loss rate and delay are discussed and presented at normalized load. Moreover, a detailed FPGA implementation is given. Practical network traffic tests provide evidence supporting the feasibility and stability of the proposed fabric. © 2014 The Institute of Electronics, Information and Communication Engineers.

关键词： Switching networks

来源：评论

学校读者我要写书评

暂无评论

Design and implementation of Load-Balanced Multipath Self-routing switching system

Design and implementation of Load-Balanced Multipath Self-ro...

引用

International MultiConference of Engineers and Computer Scientists, IMECS 2014

作者： Zhan, Qian Li, Hui Chen, Fuxing Ma, Li Peking University Shenzhen Graduate School Shenzhen GuangDong518055 China Shenzhen Eng. Lab. of Converged Networks Technology Shenzhen Key Lab. of Cloud Computing Tech. and App. China

ISBN: (纸本)9789881925336

In order to ensure high quality of service (QoS) for Next Generation Network (NGN), we construct an innovative Load-Balanced Multipath Self-routing Switching Structure which consists of the same two multipath self-routing fabrics. The result of simulation is inspiring for achieving 100% throughput and no delay or jitter. For this reason, we start on the implementation on an Altera StratixIV FPGA. And the whole FPGA system is designed into two collaborative components: the UDP system and the register system. With two algorithms around input and output two stages, incoming traffic is transformed into uniformity and then to their final destinations. During the later period debugging, software simulation platform and automated test platform are built, which contribute to our work very much. At last, we carry out several experiments to test and verify our system. The report of the test result accords with what we expected.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised hierarchical co-clustering

Semi-supervised hierarchical co-clustering

引用

7th International Conference on Rough Sets and Knowledge technology, RSKT 2012

作者： Huang, Feifei Yang, Yan Li, Tao Zhang, Jinyuan Rutayisire, Tonny Mahmood, Amjad School of Information Science and Technology Provincial Key Lab. of Cloud Computing and Intelligent Technology Southwest Jiaotong University Chengdu 610031 China School of Computer Science Florida International University Miami FL 33199 United States

ISBN: (纸本)9783642318993

Hierarchical co-clustering aims at generating dendrograms for the rows and columns of the input data matrix. The limitation of using simple hierarchical co-clustering for document clustering is that it has a lot of feature terms and documents, and it also ignores the semantic relations between feature terms. In this paper a semi-supervised clustering algorithm is proposed for hierarchical co-clustering. In the first step feature terms are clustered using a little supervised information. In the second step, the feature terms are merged as new feature attributes. And in the last step, the documents and merged feature terms are clustered using hierarchical co-clustering algorithm. Semantic information is used to measure the similarity during the hierarchical co-clustering process. Experimental results show that the proposed algorithm is effective and efficient. © 2012 Springer-Verlag.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Editorial

引用

Science China(Physics,Mechanics & Astronomy) 2014年第8期57卷 1413-1413页

作者： ZHANG Bo WU JunTing WANG BinQiang LI Hui National Digital Switching System Engineering Technological R&D Center Shenzhen Key Lab of Cloud Computing Technology and Application

The past decade has witnessed the birth of the golden age of precision cosmology. In particular, cosmic microwave back- ground （CMB） measurements have provided a powerful observational window to probe the earliest epochs of our Universe. Recently （March, 2014）, the Background Imaging of Cosmic Extragalactic Polarization （BICEP2） telescope at the South Pole announced the first detection of primordial B-mode polarization of the CMB that, if confirmed,

关键词：编辑宇宙微波背景宇宙学观察窗口河外星系 CMB 望远镜影像

来源：评论

学校读者我要写书评

暂无评论

G-MAP: General Memory-Augmented Pre-trained Language Model for Domain Tasks

G-MAP: General Memory-Augmented Pre-trained Language Model f...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Wan, Zhongwei Yin, Yichun Zhang, Wei Shi, Jiaxin Shang, Lifeng Chen, Guangyong Jiang, Xin Liu, Qun Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Science China University of Chinese Academy of Sciences China Huawei Noah's Ark Lab Huawei Cloud Computing Zhejiang Lab China

Recently, domain-specific PLMs have been proposed to boost the task performance of specific domains (e.g., biomedical and computer science) by continuing to pre-train general PLMs with domain-specific corpora. However, this Domain-Adaptive Pre-Training (DAPT;Gururangan et al. (2020)) tends to forget the previous general knowledge acquired by general PLMs, which leads to a catastrophic forgetting phenomenon and sub-optimal performance. To alleviate this problem, we propose a new framework of General Memory-Augmented Pre-trained Language Model (G-MAP), which augments the domain-specific PLM by a memory representation built from the frozen general PLM without losing any general knowledge. Specifically, we propose a new memory-augmented layer, and based on it, different augmented strategies are explored to build the memory representation and then adaptively fuse it into the domain-specific PLM. We demonstrate the effectiveness of G-MAP on various domains (biomedical and computer science publications, news, and reviews) and different kinds (text classification, QA, NER) of tasks, and the extensive results show that the proposed G-MAP can achieve SOTA results on all tasks. © 2022 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

DRWS: A model for learning distributed representations for words and sentences

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2014年 8862卷 196-207页

作者： Yan, Chunwei Zhang, Fan Huang, Lian’en The Shenzhen Key Lab for Cloud Computing Technology and Application Peking University Shenzhen Graduate School ShenzhenGuangdong China

Vector-space distributed representations of words can capture syntactic and semantic regularities in language and help learning algorithms to achieve better performance in natural language processing tasks by grouping similar words. With progress of machine learning techniques in recent years, much attention has been paid on this field. However, many NLP tasks such as text summary and sentence matching treat sentences as atomic units. In this paper, we introduce a new model called DRWS which can learn distributed representations for words and variable-length sentences. Feature vectors for words and sentences are learned based on their probability of co-occurrence between words and sentences using a neural network. To evaluate feature vectors learned by our model, we applied our model on the tasks of detecting word similarity and text summarization. Extensive experiments demonstrate the effectiveness of our proposed model in learning vector representations for words and sentences. © Springer International Publishing Switzerland 2014.

关键词： Vectors

来源：评论

学校读者我要写书评

暂无评论

A TWO-PHASE METHOD WITH IDBSCAN AND ACO FOR LOGISTICS DISTRIBUTION

A TWO-PHASE METHOD WITH IDBSCAN AND ACO FOR LOGISTICS DISTRI...

引用

2012 IEEE 2nd International Conference on cloud computing and Intelligence Systems

作者： Hongfeng Zhang Yan Yang Hongjun Wang Fei Teng School of Information Science ＆ Technology Provincial Key Lab of Cloud Computing and Intelligent TechnologySouthwest Jiaotong University

The difficulties in logistics distribution lie in the determination of distribution routes. That can be reduced to VRP （Vehicle Routing Problems）, which was raised and focused by scholars in 1950s and lots of methods have been applied to. One of the effective solutions is to divide VRP into independent TSP （Traveling Salesman Problems） and then solve them separately, which is a typical two-phase method. In the first phase, the division of VRP will be realized by IDBSCAN （Improved Density-Based Spatial Clustering of Application with Noise）, which is extended from DBSCAN （Density-Based Spatial Clustering of Applications with Noise）. In the second stage, all TSP in clusters will be solved by Ant Colony Optimization （ACO） algorithm and the solution set is the ultimate distribution routes, then experiment in the data set which is a quarter of sales data from E-commerce site company. For better simulation, the distance between two customer sites in the sales data will be measured by practical driving distance. This improved algorithm is experimentally turned out to be effective and competitive.

关键词： Clustering analysis Ant colony optimization （ACO） Vehicle routing problems （VRP）

来源：评论

学校读者我要写书评

暂无评论

Online Midgress-Sensitive Traffic Allocation for Percentile Charging in Pracitcal CDNs 31

Online Midgress-Sensitive Traffic Allocation for Percentile ...

引用

31st IEEE/ACM International Symposium on Quality of Service, IWQoS 2023

作者： Zhan, Huiyou Tan, Haisheng Xu, Huang Zhang, Chi Ni, Hongqiu Zhang, Pengfei Shan, Weihua Li, Xiang-Yang University of Science and Technology of China Cas Key Lab of Wireless-Optical Communications China USTC-Deqing Alpha Innovation Research Institute Zhejiang China Huawei Cloud Computing Technologies Co. Ltd. Algorithm Innovation Lab Xi'an China

ISBN: (纸本)9798350399738

The traffic bandwidth costs comprise a significant amount of operating expenditure in CDNs, induced by the traffic from the end-users to edge servers (edge cost) and from the edge to the center servers (midgress cost). Traffic allocation is the main approach to minimizing the total bandwidth cost. The joint optimization of the total costs is challenging, specifically when the percentile charging mechanism as well as some other practical issues are considered, such as the dynamicity of midgress traffic and the granularity of traffic allocation. In this work, based on our novel miss ratio prediction mechanism, we propose the first online framework, named Iris, jointly optimizing the edge and midgress costs under the 95th percentile charging in commercial CDNs. Iris can theoretically achieve a competitive ratio of 1+pe pc, when the miss ratio of all domains is set as b. Here pe and pc are the unit bandwidth price of the edge and midgress cost, respectively. Iris is tolerant to prediction errors which can be deployed in practical CDN systems. Extensive experiments based on real data indicate that Iris can dramatically reduce bandwidth costs by about 8.149% compared with the SOTA schemes, potentially saving millions of dollars per month for our large-scale commercial CDN collaborator. © 2023 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

A smart low-consumption IoT framework for location tracking and its real application

A smart low-consumption IoT framework for location tracking ...

引用

International Conference on Electronics Information and Emergency Communication

作者： Hao Tang Jian Shi Kai Lei Shenzhen Key Lab for Cloud Computing Technology & Applications (SPCCTA) Graduate School Shenzhen Guangdong China

ISBN: (纸本)9781509019984

Location-based services are increasingly popular. Smart devices embedded with GPS receiver are widely used in vehicle tracking, law enforcement and person monitoring. To provide real-time data collecting and analyzing, a backend server is necessary. A mobile app can work as a user client for accessing location data and remote control. In this paper, we describe the design and implementation of a system providing real-time positioning and tracking service. Our system consists of three parts: tracking device, backend server and mobile app. Our tracking device integrates GPS module, GSM/GPRS module and TCP/IP stack, capable of generating location data, establishing TCP connection and receiving SMS. Location data are transmitted over TCP connection from device to our backend server. The server is responsible for collecting and storing data. The server also provides an HTTP-based RESTful API for mobile app and web service. A mobile app is built for displaying data and controlling device remotely. Data accessing is via calling REST API and remote control is sent to device by SMS. Our device is compact in volume, accurate in positioning, and long in battery life. Our device has been integrated in a walking stick as a real-world product.

关键词： Servers Mobile communication Global Positioning System Legged locomotion Computer architecture Real-time systems Ground penetrating radar

来源：评论

学校读者我要写书评

暂无评论

A general co/decoder of network coding in HDL

A general co/decoder of network coding in HDL

引用

IEEE International Symposium on Network Coding

作者： Zhang, Minglong Li, Hui Chen, Fuxing Hou, Hanxu An, Huiyao Wang, Wei Huang, Jiaqing Shenzhen Key Lab. of Cloud Computing Tech. and Application Shenzhen Graduate School Peking Univ. 518055 China Dept. of Electronic and Information Engineering Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)9781612841403

This paper presents a practical and general coder and decoder of network coding (NC) with HDL (Hardware Description Language) logic for wire-speed nodes in multisource multicast networks. The NC coders apply random linear network coding (RLNC) and the decoders recover the original packets by Cramer's rule. All these mathematical operations are carried out in the Galois Field (2 8). The structures and algorithms of NC coder and decoder were designed in detail and implemented in HDL with NetFPGA boards provided by Stanford University. Comparing with traditional stored-and-forward mechanism, network emulations showed that networks with wire-speed NC coder and decoder nodes could achieve the capacity bound of max-flow min-cut theorem in case of bottlenecks, and the end-to-end delay was guaranteed on a small constant. © 2011 IEEE.

关键词： Network coding

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：