Gnutella-like Peer-to-Peer Networks, due to their extreme connectivity fluctuations, are highly robust against random failures. However, such error tolerance comes at a high price of attack survivability. In this pape...
详细信息
ISBN:
(纸本)0769524052
Gnutella-like Peer-to-Peer Networks, due to their extreme connectivity fluctuations, are highly robust against random failures. However, such error tolerance comes at a high price of attack survivability. In this paper, to enhance such attack survivability, we propose a new formulation used for defense against deliberately attacks based on two leading concepts: cost and load the cost measures how expensive it is to cure an attacked or infected node, and the load measures how important a link between two nodes is when propagating attacks or updating immunization information reversely. the combination of these factors leads us to introduce the concept of optimal targeted immunization, which formalizes the ideas of minimizing the risk of epidemic outbreaks in these networks. Using this analysis framework, we then devise a novel efficient targeted immunization scheme. the simulation results under a realistic Gnutella network show that our immunization scheme outperforms other existing methods, producing an arresting increase of the network attack tolerance at a lower price of eliminating malicious attacks.
distributedapplications operating in pervasive computing environment should have abilities to cope withthe high heterogeneity and variability of this environment. In a high abstract level, the dynamic software archi...
详细信息
Traditional loop-scheduling schemes are designed for homogeneous computing environments, and are probably not suitable for emerging heterogeneous PC clusters. this paper applies a two-phased method, named HPLS (Hybrid...
详细信息
ISBN:
(纸本)0769524052
Traditional loop-scheduling schemes are designed for homogeneous computing environments, and are probably not suitable for emerging heterogeneous PC clusters. this paper applies a two-phased method, named HPLS (Hybrid parallel Loop Scheduling), to heterogeneous PC-cluster environments. the key idea is to distribute most of the workload to each node according to its performance, which is modeled and estimated in advance. Experiments on our testbed PC-cluster showed that, in most cases, our method could reduce the execution time of application programs more than previous schemes could.
View-Oriented parallel Programming is based on distributed Shared Memory which is friendly and easy for programmers to use. It requires the programmer to divide shared data into views according to the memory access pa...
详细信息
Configuration for products in the distributed and dynamic computing environment motivates the demand for semantic based cooperation. Stand-alone data model of configuration systems does not meet the new requirements. ...
详细信息
this paper is aims at heterogeneity between local three-dimensional body scanning database, presents a Sort of data sharing, publishing and retrieving system architecture and approach over heterogeneous three-dimensio...
详细信息
ISBN:
(纸本)0769524052
this paper is aims at heterogeneity between local three-dimensional body scanning database, presents a Sort of data sharing, publishing and retrieving system architecture and approach over heterogeneous three-dimensional body scanning data sets distributed over Internet. the system mainly involve three-dimensional body scanning and apparel domain ontology model, domain rule database, an intelligent and semantic query engine, query facts database, analysis tools, maintenance and interface components etc. We constructed our three-dimensional body scanning database system and produced a standard model in our sharing web,it system according its ontology model. the proposed system can effective overcome the structure ana semantic heterogeneity of local three-dimensional body scanning database.
In recent years, distributed 3D sights simulation based on internet is attracting more and more attention for its many practical application. Based on Web Service technology, we design and develop a distributed simula...
详细信息
the dynamical Grid environment requires flexible and adaptive workflow system. this paper integrates Agents and Web Services into Grid Service Workflow System based on Colored Petri Nets. the combination of Agents and...
详细信息
the proceedings contain 21 papers from the Reliable Software technology - Ada-Eurpe 2005: 10th Ada-Europe internationalconference on Reliable Software technologies. the topics discussed include: ILTIS - the legacy of...
详细信息
the proceedings contain 21 papers from the Reliable Software technology - Ada-Eurpe 2005: 10th Ada-Europe internationalconference on Reliable Software technologies. the topics discussed include: ILTIS - the legacy of a success product;an Ada framework for QoS-aware applications;efficient alternatives for implementing fixed-priority schedulers;using the AADL to describe distributedapplications from middleware to software components;smart certification of mixed criticality systems;observing the development of a reliable embedded system;the implementation of Ada 2005 interface type in the GNAT compiler;and integrating application-defined scheduling withthe new dispatching policies for Ada tasks.
Component-based software development has become a popular paradigm in software engineering. From the theoretical point of view, components can be seen as coalgebras. In this paper, we present a component specification...
详细信息
暂无评论