The proceedings contain 93 papers. The topics discussed include: the anatomy of the grid: enabling scalable virtual organizations;programming high performance applications in grid environments;grid RPC meets data grid...
ISBN:
(纸本)0769510108
The proceedings contain 93 papers. The topics discussed include: the anatomy of the grid: enabling scalable virtual organizations;programming high performance applications in grid environments;grid RPC meets data grid: network enabled services for data farming on the grid;the world wide computer: prospects for parallel and distributed computing on the web;MPIFT: architecture and taxonomies for fault-tolerant, message-passing middleware for performance-portable parallel computing;effective Internet gridcomputing for industrial users;design of a generic platform for efficient and scalable clustercomputing based on middleware techniques;optimizing execution of component-based applications using group instances;containers: a sound basis for a true single system image;QoS-aware discovery of wide-area distributed services;and user preference driven multiobjective resource management in grid environments.
Hypergraph Neural Networks (HGNNs) are increasingly utilized to analyze complex inter-entity relationships. Traditional HGNN systems, based on a hyperedge-centric dataflow model, independently process aggregation task...
详细信息
This paper describes a tool that mimics the design of the remote procedure call (RPC) system to support the building of replicated objects in a cluster of workstations (COW). The tool includes an interface definition ...
详细信息
ISBN:
(纸本)0769510116
This paper describes a tool that mimics the design of the remote procedure call (RPC) system to support the building of replicated objects in a cluster of workstations (COW). The tool includes an interface definition language for describing a replica group, a language preprocessor and a runtime library system. The paper also presents one example and discusses some performance issues for replicated objects.
TACO (Topologies and Collections) is a template library that introduces the flavour of distributed data parallel processing by means of reusable topology classes and C++ templates. This paper introduces TACO's bas...
详细信息
ISBN:
(纸本)0769510108;0769510116
TACO (Topologies and Collections) is a template library that introduces the flavour of distributed data parallel processing by means of reusable topology classes and C++ templates. This paper introduces TACO's basic abstractions and provides a performance analysis for basic collective operations on various cluster architectures with several different networks.
We have developed JaWS, a Java-bused web computing system, which enables users to effortlessly export their machines in a global market of processing capacity to host remote computations [6]. Leases are used to promot...
详细信息
ISBN:
(纸本)0769510116
We have developed JaWS, a Java-bused web computing system, which enables users to effortlessly export their machines in a global market of processing capacity to host remote computations [6]. Leases are used to promote dynamic task placement as well as fair compensation for the host providers. In this paper, ive present an updated protocol used by hosts and applications to interact with the JaWS market, through which resource allocation takes place. Although this work is carried out in the context of JaWS, it can also be applied to other market-based resource allocation frameworks.
Resource management is an important infrastructure in the gridcomputing environment. Scalability and adaptability are two key challenges in the implementation of such complex software systems. In this work we introdu...
详细信息
ISBN:
(纸本)0769510108;0769510116
Resource management is an important infrastructure in the gridcomputing environment. Scalability and adaptability are two key challenges in the implementation of such complex software systems. In this work we introduce a new model for resource management in a metacomputing environment using a hierarchy of homogeneous agents that has the capability of service discovery. The performance of the agent system can be improved using different combinations of optimisation strategies. A modelling and simulation environment has been developed in this work that enables the performance of the system to be investigated. A simplified model of the resource management infrastructure is given as a case study and simulation results are included that show the impact of the choice of performance optimisation strategies on the overall system performance.
BIP [5] is a low level communication layer for the Myrinet network which includes a dedicated fi rm wa re. We identified several flaws in the design of this firmware and based on our experience decided to rethink it. ...
详细信息
ISBN:
(纸本)0769510116
BIP [5] is a low level communication layer for the Myrinet network which includes a dedicated fi rm wa re. We identified several flaws in the design of this firmware and based on our experience decided to rethink it. Our goal is to better use the hardware resources and to ensure the best possible flows of data through the channel controlled by the firmware. This results in new architecture based on an event-driven heart, optimized pipelining of communications and the possibility to use hardware communication channels asynchronously. Experiments demonstrate the efficiency of the new design.
暂无评论