We introduce analytical models for predicting the performance of World Wide Web (WWW) proxy caches that are distributed over multiple proxy machines under various management protocol assumptions. The purpose of these ...
详细信息
We introduce analytical models for predicting the performance of World Wide Web (WWW) proxy caches that are distributed over multiple proxy machines under various management protocol assumptions. The purpose of these models is to determine which distributed cache management (DCM) protocols are to be used for large-scale distributed caches. The accuracy and usefulness of our models is assessed by comparing the performance predicted by the models with measurements from real system in operation. We run tests with different values of various parameters that define behavior of distributed cache, communication protocol, DCM protocol, and application.
The increasing popularity of information services that rely on content delivery in mobile environments motivates the need for a mobile push service - an efficient and flexible content dissemination service that target...
详细信息
The increasing popularity of information services that rely on content delivery in mobile environments motivates the need for a mobile push service - an efficient and flexible content dissemination service that targets mobile users. We analyze the features of a mobile push service by investigating representative usage scenarios and propose an architecture for mobile content delivery systems. The architecture is based on the publish/subscribe (P/S) paradigm which supports many-to-many interaction of loosely-coupled entities. We define the set of services that need to collaborate with the P/S infrastructure to address the dynamics of mobile environments.
Ubiquitous computing is poised to revolutionize the way we compute and interact with each other. However, unless privacy concerns are taken into account early in the design process, we will end up creating a very effe...
详细信息
Ubiquitous computing is poised to revolutionize the way we compute and interact with each other. However, unless privacy concerns are taken into account early in the design process, we will end up creating a very effective distributed surveillance system, which would be a dream come true for electronic stalkers and "big brothers". We present a protocol, which preserves the privacy of users and keeps their communication anonymous. In effect, we create a "mist" that conceals users from the system and other users. Yet, users will still be able to enjoy seamless interaction with services and other entities that wander within the ubiquitous computing environment.
We present a method of timestamping messages and events in synchronous computations that capture the order relationship with vectors of size less than or equal to the size of the vertex cover of the communication topo...
详细信息
We present a method of timestamping messages and events in synchronous computations that capture the order relationship with vectors of size less than or equal to the size of the vertex cover of the communication topology of the system. Our method is fundamentally different from the techniques of Fidge (1989) and Mattern (1989). The timestamps in our method do not use one component per process but still guarantee that the order relationship is captured accurately. Our algorithm is online and only requires piggybacking of timestamps on program messages. It is applicable to all programs that either use programming languages based on synchronous communication such as CSP or use synchronous remote procedure calls.
This paper envisions a shared virtual reality system that could handle millions of users and objects. The SOLIPSIS system does not rely on servers and is based on a network of peers that collaborate to build up a comm...
详细信息
This paper envisions a shared virtual reality system that could handle millions of users and objects. The SOLIPSIS system does not rely on servers and is based on a network of peers that collaborate to build up a common virtual world. Real-time interactions and virtual co-presence are enabled by the means of emissive and receptive fields that determine how avatar kinestesia and multimedia streams are established. We describe here the overall architecture and the main algorithms that make Soupsis conceivable.
As the peer-to-peer (P2P) computing paradigm receives more and more attention, it is now necessary to provide computing platforms to offer an environment to build P2P applications. In this paper, we describe Rheeve, a...
详细信息
As the peer-to-peer (P2P) computing paradigm receives more and more attention, it is now necessary to provide computing platforms to offer an environment to build P2P applications. In this paper, we describe Rheeve, a computing platform that provides support for the development of HTTP-enabled, efficient and fault-tolerant P2P applications. Rheeve has a highly modular system architecture and scalable mechanisms for efficient peer connections and service discovery and delivery. Its work distribution and execution facility enables sharing of computation resources in a language and protocol independent manner. Rheeve also provides a visual programming interface, which, together with the underlying JavaBeans technology, supports a plug-n-play style of application design.
Research on aspect oriented programming and specification has highlighted the need to deal with cross-cutting concerns that involve more than one implementation level component. distributed systems are an important ap...
详细信息
Research on aspect oriented programming and specification has highlighted the need to deal with cross-cutting concerns that involve more than one implementation level component. distributed systems are an important application area where cross-cutting concerns frequently emerge. We present a novel way of composing distributed behavior from reusable superposition steps. The steps are given in a joint action notation, which allows the introducing of logically related data and operations in several implementation level components simultaneously. A superposition step contains a description of the context in which it is applicable, and the additional structure to add to a specification. Aspects of collective behavior may overlap at the implementation level. To make it possible to separate aspects cleanly at the specification level, our specification method allows entities to be merged in composition to allow an implementation level entity to play multiple specification level roles.
This paper presents in detail an efficient and provably correct algorithm for database replication over partitionable networks. Our algorithm avoids the need for end-to-end acknowledgments for each action while suppor...
详细信息
This paper presents in detail an efficient and provably correct algorithm for database replication over partitionable networks. Our algorithm avoids the need for end-to-end acknowledgments for each action while supporting network partitions and merges and allowing dynamic instantiation of new replicas. One round of end-to-end acknowledgments is required only upon a membership change event such as a network partition. New actions may be introduced to the system at any point, not only while in a primary component. We show how performance can be further improved for applications that allow relaxation of consistency requirements. We provide experimental results that demonstrate the efficiency of our approach.
Due to the increasing number of registered trademarks, it is more and more difficult to design and register a new trademark without a good retrieval method Thus, we propose a new region based shape description method:...
详细信息
Due to the increasing number of registered trademarks, it is more and more difficult to design and register a new trademark without a good retrieval method Thus, we propose a new region based shape description method: the distance-angle pair wise histogram, as the trademark shape feature vector to retrieve similar trademarks Experiments have been conducted on about 1000 trademark images. Some moment-based methods have also been implemented. In general the proposed method performs better than other moment-based methods.
The on-demand downloading of application code over a wireless network is an emerging paradigm for the distributed environment of mobile computing devices. In this environment, a user must wait while a mobile computing...
详细信息
The on-demand downloading of application code over a wireless network is an emerging paradigm for the distributed environment of mobile computing devices. In this environment, a user must wait while a mobile computing device fetches the application code over the network prior to execution. To reduce this downloading latency, we have developed a download approach based on working sets, and have designed and implemented a working-set generator that defines working sets by focusing on the intervals during which a program refers to junctions, variables, and constants. The program invocation times with this approach, simulated using trace information obtained from actual program behavior on Linux, have been compared with those of all-at-once, page, class, and symbol download approaches. This simulation showed that the working-set approach can reduce the download-execution time compared to that with the four other approaches.
暂无评论