The notion of intrinsic references, i.e. references based on the hash digest of the referent, is introduced and contrasted with that of physical references, where the referent is defined relative to the state of a phy...
详细信息
The notion of intrinsic references, i.e. references based on the hash digest of the referent, is introduced and contrasted with that of physical references, where the referent is defined relative to the state of a physical system. A retrieval mechanism using intrinsic references, the Elephant Store, is presented. The use of intrinsic references in hierarchical data structures is discussed, and the advantages regarding version management, consistency anddistributed storage are argued.
This paper presents the design and implementation of a continuous media file system, which has been implemented in the context of a distributed multimedia application development environment that has been prototyped. ...
详细信息
This paper presents the design and implementation of a continuous media file system, which has been implemented in the context of a distributed multimedia application development environment that has been prototyped. To make a performance analysis of file systems anddistributed object services for continuous media provisioning, we validate the performance analysis of the file system on media-on-demand (MOD) systems against that of a conventional file system through an experimental evaluation.
We present the first snap-stabilizing propagation of information with feedback (PIF) protocol in arbitrary networks. A snap-stabilizing protocol, starting from any arbitrary initial system configuration, always behave...
详细信息
We present the first snap-stabilizing propagation of information with feedback (PIF) protocol in arbitrary networks. A snap-stabilizing protocol, starting from any arbitrary initial system configuration, always behaves according to its specification. Our protocol is distributed, deterministic, and does not use a pre-constructed spanning tree.
We propose a fully distributed system (as compared to centralized and partially distributed systems) for cost-sensitive data mining. Experimental results have shown that this approach achieves higher accuracy than bot...
详细信息
We propose a fully distributed system (as compared to centralized and partially distributed systems) for cost-sensitive data mining. Experimental results have shown that this approach achieves higher accuracy than both the centralized and partially distributed learning methods, however, it incurs much less training time, neither communication nor computation overhead.
Data replication in distributed databases has been investigated extensively with the hope that it will improve performance, reliability, and availability. However the growth of the Internet has shown us that current r...
详细信息
Data replication in distributed databases has been investigated extensively with the hope that it will improve performance, reliability, and availability. However the growth of the Internet has shown us that current replica management do not work well when the replicas are connected by an unreliable network, subject to congestion and dynamic topology changes. In this paper we present a replica update protocol that handles an adaptive partial replication scheme on such a network.
This paper presents a survey of existing event systems structured as a taxonomy of distributed event-based programming systems. Our taxonomy identifies a set of fundamental properties of event-based programming system...
详细信息
This paper presents a survey of existing event systems structured as a taxonomy of distributed event-based programming systems. Our taxonomy identifies a set of fundamental properties of event-based programming systems and categorizes them according to the event model and event service criteria. The event service is further classified according to its organization and interaction model, as well as other functional and non-functional features.
The coordination paradigm has been used extensively as a mechanism for software composition and integration. However, relatively little work has been done for the cases where the software components involved have real...
详细信息
The coordination paradigm has been used extensively as a mechanism for software composition and integration. However, relatively little work has been done for the cases where the software components involved have real-time requirements. The paper presents an extension to a state-of-the-art controller event-driven coordination language with real-time capabilities. It then illustrates the expressiveness of the proposed extensions by means of modeling a distributed multimedia application. Finally, it discusses how these extensions can be supported by the underlying architecture.
In distributed applications, a group of multiple processes are cooperating where multimedia messages are exchanged among the processes. The multimedia objects are longer than traditional messages and are structured. I...
详细信息
In distributed applications, a group of multiple processes are cooperating where multimedia messages are exchanged among the processes. The multimedia objects are longer than traditional messages and are structured. If messages transmitted in a network are causally delivered by using traditional group protocols, computation and communication overheads are increased due to the large size of multimedia data. In this paper, we discuss new types of causally precedent relations among multimedia objects transmitted in a group of multiple processes. We discuss a protocol to causally deliver multimedia objects in a group of processes.
Cooperative Search Engine (CSE) is a distributed search engine, which can update indexes in a very short time for the purpose of fresh information retrieval. In CSE, the retrieval performance is dependent on cache con...
详细信息
Cooperative Search Engine (CSE) is a distributed search engine, which can update indexes in a very short time for the purpose of fresh information retrieval. In CSE, the retrieval performance is dependent on cache contents because communication delay occurs at retrieval time. On the other hand, however, cache is invalidated as soon as indexes are updated. Therefore, we need persistent cache that can hold valid data before and after updating. In this paper we describe the principle and evaluations of persistent cache.
We focus on an important problem in ubiquitous computing, namely, programming support for the distributed heterogeneous computing elements that make up this environment. We address the interactive, dynamic, and stream...
详细信息
We focus on an important problem in ubiquitous computing, namely, programming support for the distributed heterogeneous computing elements that make up this environment. We address the interactive, dynamic, and stream-oriented nature of this application class and develop appropriate computational abstractions in the D-Stampede distributed programming system. The key features of D-Stampede include indexing data streams temporally, correlating different data streams temporally, performing automatic distributed garbage collection of unnecessary stream data, supporting high performance by exploiting hardware parallelism where available, supporting platform and language heterogeneity, and dealing with application level dynamism. We discuss the features of D-Stampede, the programming ease it affords, and its performance.
暂无评论