检索结果-内蒙古大学图书馆

ChinaGrid Annual Conference (ChinaGrid)

作者： Ke Yang Jianhua Gu Tianhai Zhao Guofei Sun High Performance Computing Center School of Computer Science and Technology Northwestern Polytechnical University Xian China

The migration technology of virtual machine has received extensive attention in load balancing. In this paper, we purpose an optimized control strategy which combines multi-strategy mechanism with the prediction mechanism. According to the weighted average of the CPU, memory, I/O and network bandwidth utilization, we divide the hosts into four status domains. The hosts within different load status domain adopt different migration strategy, which considers migration timing, migration candidate VM and migration destination. Through this strategy, it reduces the number of the overloaded hosts, avoids instantaneous peak problem caused by the migration of virtual machines, solves the imbalance problem and the high-cost problem in tradition scheduling algorithm of migration. Experimental results demonstrate that this strategy is able to reduce the migration times and improve system performance.

关键词： Virtual machining Bandwidth Monitoring Memory management Load management Timing Load modeling

来源：评论

学校读者我要写书评

暂无评论

面向同时多线程的可扩展高速缓存动态划分方法

面向同时多线程的可扩展高速缓存动态划分方法

引用

2012全国高性能计算学术年会

作者： WU Jun-Min 吴俊敏 ZHU Xiao-Dong 朱小东 SUI Xiu-Feng 隋秀峰 JIN Ying-Qi 金瑛棋 ZHAO Xiao-Yu 赵小雨 School of Computer Science and Technology University of Science and Technology of China Hefei 230 中国科学技术大学计算机学院合肥230027 中国科学技术大学苏州研究院院苏州215123 National Research Center for Intelligent Computing Systems Institute of Computing Technology Beiji 中国科学院计算技术研究所国家智能计算研究中心北京100091

提出一种面向大规模同时多线程处理器可扩展高速缓存的动态划分方法.将文献所提的机制扩展到多体高速缓存中.线程的内存访问特征变化迅速，因此在运行时收集同时多线程的缺失率特征，依此特征在线程间划分高速缓存.实验结果显示，相对... 详细信息

提出一种面向大规模同时多线程处理器可扩展高速缓存的动态划分方法.将文献所提的机制扩展到多体高速缓存中.线程的内存访问特征变化迅速，因此在运行时收集同时多线程的缺失率特征，依此特征在线程间划分高速缓存.实验结果显示，相对于非划分的具有标准LRU替换策略的高速缓存，本文提出的划分方法能够IPC达18.94％.

关键词：多线程处理器高速缓存动态划分

来源：评论

学校读者我要写书评

暂无评论

A Classification Framework of Service Composition Issues in BIRIS Scenario

A Classification Framework of Service Composition Issues in ...

引用

International Joint Conference on Service sciences (IJCSS)

作者： Dianhui Chu Zhongjie Wang Xiaofei Xu Research Center of Intelligent Computing for Enterprises and Services (ICES) School of Computer Science and Technology Harbin Institute of Technology Harbin China

Service composition (SC) is currently one of the most popular research topics in service computing domain. Given a specific functional and non-functional requirements on a specific service, a composition algorithm selects a set of service components from an existing repository and composes them together as a total solution to fulfill the requirement. There have been fruitful achievements, e.g., template-based composition, planning-based composition, automated/semi-automated/manual composition, etc. However, along with BIRIS becomes a dominate form of modern service systems, a series of new appearing features bring great challenges to traditional SC researches, e.g., (1) from one single requirement to temporally sequential requirements, (2) the available services from abundant to limited, (3) from orchestration-oriented composition solution to choreography-oriented ones, (4) from specified requirements to ambiguous and personalized ones, (5) from the objective of maximizing customer satisfaction to ensuring the benefits of customers and providers together, etc. This paper analyzes these challenges elaborately and presents a classification framework which includes 4 dimensions and 13 sub-dimensions. Several typical SC issues in BIRIS scenario are identified under the framework.

关键词： Web services Optimization Quality of service Customer satisfaction Manuals

来源：评论

学校读者我要写书评

暂无评论

The Quantitative Evaluation on the Value-Oriented Priority of Service Elements

The Quantitative Evaluation on the Value-Oriented Priority o...

引用

International Joint Conference on Service sciences (IJCSS)

作者： Chao Ma Xiaofei Xu Zhongjie Wang Research Center of Intelligent Computing for Enterprises and Services (ICES) School of Computer Science and Technology Harbin Institute of Technology Harbin China

Value is considered as the ultimate goal of a good service, and whether and to what degree the expected value could be delivered to customers and providers depends on capacities and quality of various functional service elements in service system. Different elements have different contributions to the value production. Due to the limit of available service components to construct the service system, it is necessary to measure the priority of each service element in service models so that those elements with higher value-oriented priorities (VOP) should be implemented by concrete service components earlier and with higher capability during the service system development. This paper presents a quantitative evaluation method of VOP. Basic definition and measurement of service value are firstly introduced, and then VOP is put forward following three types of particular priorities called sensitivity points, tradeoff points and key points. Their characteristics and evaluation principles are presented in details, and the process of the quantitative evaluation method is put forward subsequently. To complement the above discussion, a case study from ocean transportation service is applied for demonstration.

关键词： Quality of service Sea measurements Sensitivity Reliability Receivers Transportation Oceans

来源：评论

学校读者我要写书评

暂无评论

Towards Providing Cloud Functionalities for Grid Users

Towards Providing Cloud Functionalities for Grid Users

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Weizhou Peng Jie Tao Lizhe Wang Holger Marten Dan Chen School of Computer Science China University of Geosciences Wuhan Hubei China Steinbuch Center of Computing Karlsruhe Institute of Technology Germany Pervasive Technology Institute Indiana University USA

Grid computing uses a job submission model that requires the users to perform a set of interactive operations for executing an application on the Grid. Grid users are therefore burdened with the tasks of understanding the basic concept of Grid computing and the details of job management. Cloud computing, on the other hand, applies a utility model that allows the user to access the underlying platform via Web services. This work brings the Cloud concept to the Grid with a result of replacing the job submission model with a service infrastructure. In this case, Grid applications are presented as Web services that can be executed on the Grid automatically without user interactions. All issues related to job management are performed by the system. The service infrastructure significantly simplifies the users' task in accessing the Grid.

关键词： Production Servers Cloud computing Portals Computational modeling

来源：评论

学校读者我要写书评

暂无评论

RB树:一种支持空间近似关键字查询的外存索引

RB树:一种支持空间近似关键字查询的外存索引

引用

第29届中国数据库学术会议

作者： Wang Jinbao 王金宝 Gao Hong 高宏 Li Jianzhong 李建中 Yang Donghua 杨东华 School of Computer Science and Technology Harbin Institute of Technology Harbin 150001 哈尔滨工业大学计算机科学与技术学院哈尔滨 150001 Center for High Performance Computing The Academy of Fundamental and Interdisciplinary Sciences Ha 哈尔滨工业大学基础与交叉科学研究院高性能计算中心哈尔滨150001

空间近似关键字查询包含一个空间条件和一组关键字相似性条件，这种查询在空间数据库中返回同时满足以下条件的对象：1)对象的位置信息满足查询中的空间条件;2)对于查询中的任何一个关键字，对象中至少包含一个关键字与其相似度大于给... 详细信息

空间近似关键字查询包含一个空间条件和一组关键字相似性条件，这种查询在空间数据库中返回同时满足以下条件的对象：1)对象的位置信息满足查询中的空间条件;2)对于查询中的任何一个关键字，对象中至少包含一个关键字与其相似度大于给定阈值.随着当前数据的爆炸性增长，空间数据库无法完整地存放在内存中，因此空间数据库需要支持空间近似关键字查询的外存索引.目前，还没有在外存中支持精确的空间近似关键字查询的索引结构.设计了一种新型的外存索引RB树，在外存中支持精确的空间近似关键字查询.RB树支持的空间近似关键字查询包括多种空间条件，如范围查询、NN查询，同时支持多种关键字相似性度量，包括编辑距离、规范化编辑距离等.通过真实数据中的性能测试验证了RB树的效率.

关键词：空间数据库关键字查询处理索引结构

来源：评论

学校读者我要写书评

暂无评论

Context-specific miRNA regulation network predicts cancer prognosis

Context-specific miRNA regulation network predicts cancer pr...

引用

5th IEEE International Conference on Systems Biology, ISB 2011

作者： Zhou, Xionghui Liu, Juan Liu, Changning Rayner, Simon Liang, Fengji Ju, Jingfang Li, Yinghui Chen, Shanguang Xiong, Jianghui School of Computer Science Wuhan University Wuhan China Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China State Key Laboratory of Virology Wuhan Institute of Virology Chinese Academy of Sciences Wuhan China Department of Pathology Stony Brook University Medical Center NY United States

ISBN: (纸本)9781457716669

MicroRNAs can regulate hundreds of target genes and play a pivotal role in a broad range of biological process. However, relatively little is known about how these highly connected miRNAs-target networks are remodelled in the context of various diseases. Here we examine the dynamic alteration of context-specific miRNA regulation to determine whether modified microRNAs regulation on specific biological processes is a useful information source for predicting cancer prognosis. A new concept, Context-specific miRNA activity (CoMi activity) is introduced to describe the statistical difference between the expression level of a miRNA's target genes and non-targets genes within a given gene set (context). The microarray gene expression profile of brain tumors from 356 patients (The Cancer Genome Atlas dataset) was converted into a CoMi activity pattern, and showed significant positive correlation with the corresponding miRNA expression pattern. In a breast cancer cohort, the differential CoMi activity between good prognosis (longer survival) vs. bad prognosis patients forms a scale-free network, which highlighted a group of important cancer-related microRNAs and GO terms, e.g. hsa-miR-34a and 'cell adhesion'. Then two breast cancer cohorts were used in outcome prediction in an independent test. Using a popular T-test feature selection method and a support vector machine (SVM) classifier with 10-fold cross-validation, the CoMi activity feature achieves an area under curve (AUC) of 0.7155, better than the AUC value of 0.6339 for feature selection based on mRNA expression. In an independent test, CoMi feature selection achieved an AUC of 0.6874. Survival analysis also shows signatures defined by CoMi activity was predictive of survival and superior to mRNAs signatures. In short, we have demonstrated the first interrogation of dynamic remodeling of context specific miRNAs regulation networks in cancer. The altered microRNAs regulation on specific contexts could be used to predi

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

CDRAS: An approach to dealing with Man-in-the-Middle attacks in the context of voice over IP

CDRAS: An approach to dealing with Man-in-the-Middle attacks...

引用

2011 6th International Conference on Availability, Reliability and Security, ARES 2011

作者： Hofbauer, Stefan Quirchmayr, Gerald Wills, Christopher C. Danube Data Center GmbH Network Department Floridsdorfer Hauptstrasse 1 1210 Vienna Austria University of Vienna Multimedia Information Systems Research Group Liebiggasse 4/3-4 1010 Vienna Austria University of South Australia Division of Information Technology School of Computer and Information Science Mawson Lakes SA 5095 Australia University of South Australia Division of Information Technology School of Computer and Information Science GPO Box 2471 Adelaide SA 5001 Australia Kingston University Faculty of Computing Information Systems and Mathematics Penrhyn Road Surrey KT1 2EE London United Kingdom

ISBN: (纸本)9780769544854

This paper describes the CDRAS (Call Detail Records Analysis System) system, the motivation behind it, its approach and its background. The system aims at dealing with the notorious Man-in-the-Middle attack in the context of VoIP. As the use of such a system constitutes leads to an interference with the privacy of users, the most relevant legal aspects related to this work are also briefly discussed. © 2011 IEEE.

关键词： Voice/data communication systems

来源：评论

学校读者我要写书评

暂无评论

Traffic-Aware Spectrum Assignment in Wireless Networks

Traffic-Aware Spectrum Assignment in Wireless Networks

引用

International Conference on Networking, Architecture, and Storage (NAS)

作者： Zhang Long Pan Dashan Li Yang Fang Tao Liang Chao Li Li Lenovo Mobile Communication Technology Limited Beijing China Jiangnan Institute of Computer Technology Wuxi China School of Computer Engineer & Science High Performance Computing Center Shanghai University China School of Adult Education Henan University of Economics and Law Zhengzhou China National University of Defense Technology Changsha China

Radio frequency spectrum is a finite and scarce resource. However, past research efforts usually assume orthogonal channels of fixed bandwidth, prohibiting the effective use of the spectrum. Recently some work began to explore dynamic channel width adaptation. In this paper, by reducing the optimal channel width adaptation into the "elastic packing" problem, we devise distributed algorithms based on graph multi-interval coloring and max-coloring. We then integrate our algorithms into the multi-radio multi-channel wireless mesh networks. Simulation results show that our schemes can improve network performance significantly.

关键词： Adaptation models Logic gates Color Partitioning algorithms Telecommunication traffic Distributed algorithms Vegetation

来源：评论

学校读者我要写书评

暂无评论

The performance of GRAPE-DR for dense matrix operations

引用

Procedia computer science 2011年 4卷 888-897页

作者： Junichiro Makino Hiroshi Daisaka Toshiyuki Fukushige Yutaka Sugawara Mary Inaba Kei Hiraki Center for Computational Astrophysics National Astronomical Observatory of Japan Mitaka Tokyo 181-8588 Faculty of Commerce and Management Hitotsubashi University Kunitachi Tokyo 186-8601 K&F Computing Research Chofu Tokyo 182-0026 Department of Computer Science Graduate School of Information Science and Technology The University of Tokyo Tokyo 133-0033

We describe the implementation and performance of dense matrix multiplication and LU decomposition on the GRAPE-DR SIMD accelerator board. A GRAPE-DR card, with 4 GRAPE-DR chips, has the theoretical peak DP performance of 819 Gflops. Each GRAPE-DR chip has 512 processing elements and operates with 400 MHz clock cycle. each PE can perform one addition and one multiplication in every two clock cycles. The measured performance of matrix multiplication is 730 Gflops for the multiplication of matrices with size 51200 by 2048 and 2048 by 51200. The performance of LU decomposition is 480 Gflops for the problem size of 51200.

关键词： Numerical methods 02.60.-x

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：