检索结果-南通市图书馆

IoT Intelligence Empowered by End-Edge-Cloud Orchestration

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China Communications 2022年第7期19卷 152-156页

作者： yaoxue zhang Feng Lyu Peng Yang Wen Wu Jie Gao Department of Computer Science and Technology at Tsinghua University China School of Computer Science and Engineering Central South UniversityChangshaChina School of Electronic Information and Communications HUST Frontier Research Center Peng Cheng LaboratoryShenzhenChina 渥太华卡尔顿大学信息技术学院

To support intelligent Internet of Things(IoT)applications,such as autonomous driving,smart city surveillance,and virtual reality(VR)/augmented reality(AR),cloud services are expected to be pushed to the proximity of IoT devices for quality *** instance,to facilitate safe autonomous driving,the service delay of most vehicular applications is required to be within milliseconds,and any information delay may result in dangerous on-road conditions.

关键词： IoT delay driving

Transparent Computing: Spatio-Temporal Extension on von Neumann Architecture for Cloud Services

在线全文

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2013年第1期18卷 10-21页

作者： yaoxue zhang Yuezhi Zhou the Department of Computer Science and Technology Tsinghua University the the School of Information Science and Engineering Central South University

The rapid advancements in hardware, software, and computer networks have facilitated the shift of the computing paradigm from mainframe to cloud computing, in which users can get their desired services anytime, anywhere, and by any means. However, cloud computing also presents many challenges, one of which is the difficulty in allowing users to freely obtain desired services, such as heterogeneous OSes and applications, via different light-weight devices. We have proposed a new paradigm by spatio-temporally extending the von Neumann architecture, called transparent computing, to centrally store and manage the commodity programs including OS codes, while streaming them to be run in non-state clients. This leads to a service-centric computing environment, in which users can select the desired services on demand, without concern for these services＇ administration, such as their installation, maintenance, management, and upgrade. In this paper, we introduce a novel concept, namely Meta OS, to support such program streaming through a distributed 4VP~ platform. Based on this platform, a pilot system has been implemented, which supports Windows and Linux environments. We verify the effectiveness of the platform through both real deployments and testbed experiments. The evaluation results suggest that the 4VP~ platform is a feasible and promising solution for the future computing infrastructure for cloud services.

关键词： transparent computing extended von Neumann architecture cloud computing cloud services Meta OS

Transparent Computing: Development and Current Status

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第5期29卷 793-811页

作者： zhang yaoxue DUAN Sijing zhang Deyu REN Ju Department of Computer Science and Technology Tsinghua University School of Computer Science and Engineering Central South University

Current terminal devices, such as Personal computers(PCs), diskless workstation, mobile terminals and embedded Internet of things(IoT) devices, are facing several challenges, including limited computing capabilities and storage resources, complex maintenance and management of the software and data, as well as secure service provisioning. Facing these challenges, Cloud computing(CC) and Edge computing(EC) encourage the terminal devices to shift compute-intensive works to the cloud servers or edge nodes. Different from CC and EC, Transparent computing(TC) is a novel user-centric computing paradigm in which the hardware and software are separated in different places. Specificaly, all the software, including Operating systems(OSes), applications programs and management tools, are deposited on the servers and transmitted on demand to terminal devices in a transparent way, while the computation is performed on terminals. In this article, we give a broad survey on transparent computing about the development and current status and compare the differences with CC and EC. To realize the provisioning of the cross-platform service, TC shields hardware differences of heterogeneous devices and builds standardized hardware and software interfaces. From the perspective of security, TC entails high-level security mechanisms and the Meta OS can detect malware and OS-level attacks. Specifically, we first review the development courses of network computing paradigms and the motivation of TC. Then, we present a comparison of several emerging computing paradigms and TC from three perspectives, i.e., virtualization, location for computing, and location for storage. We also show the basic architecture of TC, development process, current status, and key enabling technologies of TC for PCs and lightweight terminals. Lastly, several new challenges and future research directions are indicated to inspire continuous investigations. We believe that this survey can help readers to obtain complete

关键词： Transparent computing Network computing Survey

A Sample-Efficient Actor-Critic Algorithm for Recommendation Diversification

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第1期29卷 89-96页

作者： LI Shuang YAN Yanghui REN Ju ZHOU Yuezhi zhang yaoxue Department of Computer Science and Technology Tsinghua University

Diversifying recommendation results gains benefits from satisfying user’s existing interests as well as exploring novel information needs. Recently proposed Monte-Carlo based reinforcement learning method suffers from sample inefficiency, large variance, and even failing to perform well in large action space. We propose a novel actor-critic reinforcement learning algorithm for recommendation diversification in order to solve the above mentioned problems. The actor acts as the ranking policy, while the introduced critic predicts the expected future rewards of each candidate action. The critic target is updated by full Bellman equation and the actor network is optimized using expected gradient in the whole action space. To further stabilize and improve the performance, we also add policy-filtered critic supervision loss. Experiments on MovieLens dataset well demonstrate the effectiveness of our approach over multiple competitive methods.

关键词： Recommender system Diversity Reinforcement learning Actor-critic

A Chan–Vese Model Based on the Markov Chain for Unsupervised Medical Image Segmentation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2021年第6期26卷 833-844页

作者： Quanwei Huang Yuezhi Zhou Linmi Tao Weikang Yu yaoxue zhang Li Huo Zuoxiang He Key Laboratory of Pervasive Computing(Ministry of Education)and the Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China School of Electronic and Information Engineering Beihang UniversityBeijing 100191China Department of Nuclear Medicine Peking Union Medical College HospitalBeijing 100730China School of Clinical Medicine Tsinghua Universityand also with Beijing Tsinghua Changgung HospitalBeijing 100084China

The accurate segmentation of medical images is crucial to medical care and research;however, many efficient supervised image segmentation methods require sufficient pixel level labels. Such requirement is difficult to meet in practice and even impossible in some cases, e.g., rare Pathoma images. Inspired by traditional unsupervised methods, we propose a novel Chan–Vese model based on the Markov chain for unsupervised medical image segmentation. It combines local information brought by superpixels with the global difference between the target tissue and the background. Based on the Chan–Vese model, we utilize weight maps generated by the Markov chain to model and solve the segmentation problem iteratively using the min-cut algorithm at the superpixel *** method exploits abundant boundary and local region information in segmentation and thus can handle images with intensity inhomogeneity and object sparsity. In our method, users gain the power of fine-tuning parameters to achieve satisfactory results for each segmentation. By contrast, the result from deep learning based methods is *** performance of our method is assessed by using four Computerized Tomography(CT) datasets. Experimental results show that the proposed method outperforms traditional unsupervised segmentation techniques.

关键词： medical image unsupervised segmentation Markov chain

A Survey on Emerging Computing Paradigms for Big Data

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2017年第1期26卷 1-12页

作者： zhang yaoxue REN Ju LIU Jiagang XU Chugui GUO Hui LIU Yaping School of Information Science and Engineering Central South University Research Institute of Information Security and Big Data Central South University

The explosive growth of data volume and the ever-increasing demands of data value extraction have driven us into the era of big data. The "5V"(Variety,Velocity, Volume, Value, and Veracity) characteristics of big data pose great challenges to traditional computing paradigms and motivate the emergence of new *** computing is one of the representative technologies that can perform massive-scale and complex data computing by taking advantages of virtualized resources, parallel processing and data service integration with scalable data storage. However, as we are also experiencing the revolution of Internet-of-things(IoT), the limitations of cloud computing on supporting lightweight end devices significantly impede the flourish of cloud computing at the intersection of big data and IoT era. It also promotes the urgency of proposing new computing paradigms. We provide an overview on the topic of big data, and a comprehensive survey on how cloud computing as well as its related technologies can address the challenges arisen by big ***, we analyze the disadvantages of cloud computing when big data encounters IoT, and introduce two promising computing paradigms, including fog computing and transparent computing, to support the big data services of IoT. Finally, some open challenges and future directions are summarized to foster continued research efforts into this evolving field of study.

关键词： Computing paradigm Big data Cloud computing Transparent computing Fog computing Internet-of-things(IoT)

An End-to-End QoS Control Model for Enhanced Internet

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2000年第6期15卷 497-508页

作者：张尧学王晓春顾钧 DepartmentofComputerScienceandTechnology TsinghuaUniversityBeijing100084P.R.China

This paper describes an end-to-end QoS (Quality of Services) control model for distributed multimedia computing on enhanced Internet, and gives the design and implementation of this model including hosts and routers. The architec- ture, the mathematical definition about QoS parameters, and the mapping between Integrated Service (IS) and Differentiated Service (DS) are discussed in this paper. The simulation shows that this model can improve the performance of audio streams when it is used in IPhone system.

关键词： protocol multimedia networking QoS RSVP (Resource ReSerVation Protocol) router

基于图神经网络的联合用户调度与波束成形优化算法

在线全文

学校读者我要写书评

暂无评论

通信学报 2022年第7期43卷 73-84页

作者：何世文袁军安振宇张敏黄永明张尧学中南大学计算机学院湖南长沙410083 东南大学移动通信国家重点实验室江苏南京210096 紫金山实验室江苏南京211111 湖南邮电职业技术学院信息通信学院湖南长沙410015 清华大学计算机科学与技术系北京100084

协作多点(CoMP)传输技术具有降低同频干扰和提高频谱效率的特点。对于CoMP,用户调度与波束成形是2个分别位于媒体访问接入层和物理层的基本研究问题。在考虑用户服务质量需求下,重点研究用户调度与波束成形的联合优化问题,并以网络吞吐... 详细信息

协作多点(CoMP)传输技术具有降低同频干扰和提高频谱效率的特点。对于CoMP,用户调度与波束成形是2个分别位于媒体访问接入层和物理层的基本研究问题。在考虑用户服务质量需求下,重点研究用户调度与波束成形的联合优化问题,并以网络吞吐量最大化为目标。为了克服传统优化算法计算开销大且未有效利用网络历史数据信息的问题,提出了一种基于图神经网络联合用户调度与功率分配模型,并结合波束向量的解析公式,以实现联合用户调度与波束成形优化。仿真分析表明,所提算法能够以较低的计算开销实现与传统优化算法相匹配,甚至更优的性能表现。

关键词：跨层优化图神经网络协作多点用户调度波束成形

Performance Analysis and Improvement of Storage Virtualization in an OS-Online System

在线全文

学校读者我要写书评

暂无评论

Journal of Software Engineering and Applications 2012年第11期5卷 912-922页

作者： Yuan Gao yaoxue zhang Yuezhi Zhou Department of Computer Science and Technology Tsinghua University Beijing China

An OS-online system called TransCom is based on a virtual storage system that supports heterogeneous services of the operating system and applications online. In TransCom, OS and software which run in the client are stored on the centralized servers, while computing tasks are carried out by the clients, so the server is the bottleneck of the system performance. This paper firstly analyzes the characteristics of its real usage workload and builds a queuing model to locate the system bottlenecks. According to the study, the disk is the primary bottleneck, so an optimal two-level cache arrangement policy is developed on both the server and the client, which aims to avoid most of the server disk accesses. LRU algorithm is used in the client-side cache. A cache management algorithm called Frequency-based Multi-Priority Queues (FMPQ) proposed in this paper is used in the server-side cache. Experiment results show that the appropriate cache arrangement can significantly improve the capability of the TransCom server.

关键词： Storage Virtualization Performance Analysis Cache Strategy