Performance visualization is one of the performance evaluation techniques that can be used to perform a global analysis of a system9;s behaviour, from its internal point of view. In this work, we describe how a vis...
详细信息
ISBN:
(纸本)9783642004865
Performance visualization is one of the performance evaluation techniques that can be used to perform a global analysis of a system's behaviour, from its internal point of view. In this work, we describe how a visualization tool, initially designed for classical distributed / parallel systems, has been adapted to visualize the internal behaviour of a multi-agent system. We present such an adaptation though a multi-agent application that can be considered as a typical example for analysis and performance study.
this paper presents the design of a multi-channel distributed monitoring platform, which is highly suitable to be applied into a traction supply system for urban railway transportation. In our platform are combined th...
详细信息
this paper discusses design and implementation of a scalable high performance remote sensing satellite ground processing system using a variety of advanced hardware and software application technology on performance a...
详细信息
ISBN:
(纸本)9783642019722
this paper discusses design and implementation of a scalable high performance remote sensing satellite ground processing system using a variety of advanced hardware and software application technology on performance and function. these advanced technologies include the network, parallel file system, parallel programming, job schedule, workflow management, design patterns,etc, which make performance and function of remote sensing satellite ground processing system scalable enough to fully meet the high performance processing requirement of multi-satellite, multi-tasking, massive remote sensing satellite data. the "beijing-1" satellite remote sensing ground processing system is introduced as in instance.
the ALICE experiment at CERN LHC is intensively using a PROOF cluster for fast analysis and reconstruction. the current system (CAF - CERN Analysis Facility) consists of 120 CPU cores and about 45 TB of local space. P...
详细信息
ISBN:
(纸本)9783642019692
the ALICE experiment at CERN LHC is intensively using a PROOF cluster for fast analysis and reconstruction. the current system (CAF - CERN Analysis Facility) consists of 120 CPU cores and about 45 TB of local space. PROOF enables interactive parallel processing of data distributed on clusters Of computers or multi-core machines. Subsets of selected data are automatically staged onto CAF from the Grid storage systems. However, a cluster of the size of CAF can only hold a fraction of the yearly 3 PB data accumulated by ALICE. the impracticability to store and process such data volume in one single computing centre leads to the need to extend the concept of PROOF to the Grid paradigm.
Biclustering refers to simultaneous clustering of objects and their features. Use of biclustering is gaining momentum in areas such as text mining, gene expression analysis and collaborative filtering. Due to requirem...
详细信息
ISBN:
(纸本)9781615671090
Biclustering refers to simultaneous clustering of objects and their features. Use of biclustering is gaining momentum in areas such as text mining, gene expression analysis and collaborative filtering. Due to requirements for high performance in large scale data processing applications such as Collaborative filtering in E-commerce systems and large scale genome-wide gene expression analysis in microarray experiments, a high performance prallel/distributed solution for biclustering problem is highly desirable. Recently, Ahmad et al [1] showed that Bipartite Spectral Partitioning, which is a popular technique for biclustering, can be reformulated as a graph drawing problem where objective is to minimize Hall's energy of the bipartite graph representation of the input data. they showed that optimal solution to this problem is achieved when nodes are placed at the barycenter of their neighbors. In this paper, we provide a parallel algorithm for biclustering based on this formulation. We show that parallel energy minimization using barycenter heuristic is embarrassingly parallel. the challenge is to design a bicluster identification algorithm which is scalable as well as accurate. We show that our parallel implementation is not just extremely scalable, it is comparable in accuracy as well with serial implementation. We have evaluated proposed parallel biclustering algorithm with large synthetic data sets on upto 256 processors. Experimental evaluation shows large superlinear speedups, scalability and high level of accuracy.
the field of data mining increasingly adapt methods and algorithms from advanced matrix computations, graph theory and optimization. Prominent examples are spectral clustering, non-negative matrix factorization, Princ...
详细信息
ISBN:
(纸本)9781615671090
the field of data mining increasingly adapt methods and algorithms from advanced matrix computations, graph theory and optimization. Prominent examples are spectral clustering, non-negative matrix factorization, Principal Component Analysis (PCA) and Singular Value Decomposition (SVD), graph-Laplacian based semi-supervised learning, diffusion process, etc. Graph and matrix-based methods are rapidly becoming popular and significant in data mining for the following reasons: (1) Graph and matrix based methods are amenable to vigorous analysis and benefits from the well-established knowledge in matrix computations, graph theory and optimization;(2) Compared to probabilistic and information theoretic approaches, graph and matrix based methods are fast, easy to understand and implement;and (3) Graph and matrix based methods are especially suitable for parallel and distributed-memory computers to solve large scale challenging problems such as searching and extracting patterns from the entire Web. this tutorial will present recent advances in algorithms and methods using graphs and matrices for modeling and analyzing massive, high-dimensional, and nonlinear-structured data. One main goal of the tutorial is to consolidate the recent ideas on data mining using graphs and matrices. We will summarize some open problems contained in this field and propose some future trends. We also wish to attract practitioners who seek novel ideas for applications.
DeXIN (distributed extended XQuery for data INtegration) integrates multiple, heterogeneous, highly distributed and rapidly changing web data sources in different formats, e.g. XML, RDF and relational data. DeXIN is a...
详细信息
ISBN:
(纸本)9783642028175
DeXIN (distributed extended XQuery for data INtegration) integrates multiple, heterogeneous, highly distributed and rapidly changing web data sources in different formats, e.g. XML, RDF and relational data. DeXIN is a RESTful data integration web service which integrates heterogeneous distributed data sources, including data services (DaaS - data as a service). At the heart of DeXIN is an XQuery extension that allows users/applications to execute a single query against distributed, heterogeneous web data sources or data services. In this system demo, we show how DeXIN can provide an optimized, distributed and parallel query processing and data integration at the same time.
Web service is a new service-oriented computing paradigm which poses the unique security challenges due to its inherent heterogeneity, multidomain characteristic and highly dynamic nature. A key challenge in Web servi...
详细信息
ISBN:
(纸本)9783642030949
Web service is a new service-oriented computing paradigm which poses the unique security challenges due to its inherent heterogeneity, multidomain characteristic and highly dynamic nature. A key challenge in Web services security is the design of effective access control schemes. Attribute-based access control (ABAC) is more appropriate than some other access control mechanisms, but it do not fully exploit the semantic power and reasoning capabilities of emerging web applications. So a semantic-aware attribute-based access control model (SABAC) is presented to address these issues by combining the ABAC withthe Semantic Web technologies in this paper. SABAC grants access to services based oil attributes of the related entities, and uses Shibboleth service to address the disclosure issue of the sensitive attributes. In addition, SABAC uses the Web Ontology Language (OWL) standard to represent the ontology of the resources and users and uses eXtensible Access Control Markup Language (XACML) as the policy language. It call provide administratively scalable alternative to identity-based authorization methods and provide semantic interoperability for the access control to Web services. Moreover, SABAC also separates ontology management from access management.
作者:
Heyer, ClintABB
Strateg R&D Oil Gas & Petrochem Oslo Norway
this paper describes the design and implementation of a novel decentralized publish/subscribe framework. the primary goal of the design was for a high level of end-developer and user accessibility and simplicity. Furt...
详细信息
ISBN:
(纸本)9783642028298
this paper describes the design and implementation of a novel decentralized publish/subscribe framework. the primary goal of the design was for a high level of end-developer and user accessibility and simplicity. Furthermore, it was desired to have strong support for occasionally-connected clients and support for mobile and web-based systems. Content-based event patterns can be defined using scripts, with many common script languages supported. Script-based, stateful event patterns permit rich expressiveness, simplify client development and reduce network usage. the framework also offers event persistence, caching and publisher quenching. We also describe a number of applications already built on the framework, for example publishers to support location and presence awareness and ambient visualizations of financial data.
A number of analytical models exists that capture various properties of the BitTorrent protocol. However, until now virtually all of these models have been based on the assumption that the peers in the system have hom...
详细信息
ISBN:
(纸本)9781424450664
A number of analytical models exists that capture various properties of the BitTorrent protocol. However, until now virtually all of these models have been based on the assumption that the peers in the system have homogeneous bandwidths. As this is highly unrealistic in real swarms, these models have very limited applicability. Most of all, these models implicitly ignore BitTorrent's most important property: peer selection based on the highest rate of reciprocity. As a result, these models are not suitable for understanding or predicting the properties of real BitTorrent networks. Furthermore, they are hardly of use in the design of realistic BitTorrent simulators and new P2P protocols. In this paper, we extend existing work by presenting a model of a swarm in BitTorrent where peers have arbitrary upload and download bandwidths. In our model we group peers with (roughly) the same bandwidth in classes, and then analyze the allocation of upload slots from peers in one class to peers in another class. We show that our model accurately predicts the bandwidth clustering phenomenon observed experimentally in other work, and we analyze the resulting data distribution in swarms. We validate our model with experiments using real BitTorrent clients. Our model captures the effects of BitTorrent's well-known 'tit-for-tat' mechanism in bandwidth-inhomogeneous swarms and provides an accurate mathematical description of the resulting dynamics.
暂无评论