2000 4th International Conference on Algorithms and Architectures for Parallel Processing : ICA3PP'2000, Hong Kong, December 11-13 2000 / editors, Andrzej Goscinski ... [et al.] ; sponsors, Deakin University, Australia ... [et al.].

Format:

Book

Conference/Event

Contributor:

Goscinski, Andrzej.

Deakin University.

Conference Name:

International Conference on Algorithms and Architectures for Parallel Processing (4th : 2000 Dec. 11-13 : Hong Kong)

International Conference on Algorithms and Architectures for Parallel Processing.

Language:

English

Subjects (All):

Parallel processing (Electronic computers)--Congresses.

Parallel processing (Electronic computers).

Computer algorithms--Congresses.

Computer algorithms.

Computer architecture--Congresses.

Computer architecture.

Physical Description:

1 online resource (745 p.)

Other Title:

Algorithms and architectures for parallel processing

ICA3PP'200

Place of Publication:

Singapore ; River Edge, NJ : World Scientific, c2000.

Language Note:

English

Summary:

ICA 3 PP 2000 was an important conference that brought together researchers and practitioners from academia, industry and governments to advance the knowledge of parallel and distributed computing. The proceedings constitute a well-defined set of innovative research papers in two broad areas of parallel and distributed computing: (1) architectures, algorithms and networks; (2) systems and applications.

Contents:

Preface; CONTENTS; Chapter 1 Cluster Computing; Efficient deployment of shared memory models on clusters of PC susing the SMiLEing HAMSTER approach; 1 Motivation; 2 HAMSTER: Hybrid-dsm based Adaptive and Modular Shared memory archiTEctuRe; 3 Flexibility in implementing the target programming model; 4 Experimental evaluation and results; 4.1 Raw performance and consistency management overhead; 4.2 Numerical kernel performance; 4.3 Full application performance; 5 Related work; 6 Conclusions and Outlook; Acknowledgments; PARALLEL PROGRAMMING WITH DISTRIBUTED OBJECT GROUPS - THE TACO APPROACH

Buffer ManagementDMA or PIO; Avoiding Deadlock; Local Communication; IV Performance of FMP; Latency; Bandwidth; V Conclusion; References; A SOFTWARE DEVELOPMENT METHODOLOGY TO SUPPORT DISTRIBUTED COMPUTING CLUSTERS; 1 Introduction and History; 2 PARSA; 3 Combining Shared Memory and Distributed Memory Paradigms; 4 Distributed Computing Cluster Issues; 5 Performance; 6 Conclusions and Future Work; References; Chapter 2 Interconnection Networks and Routing; TRIPWIRE: A SYNCHRONISATION PRIMITIVE FOR VIRTUAL MEMORY MAPPED COMMUNICATION; 1 Introduction; 2 The Interconnect Hardware

2.1 Non-coherent shared memory2.2 Wire protocol; 2.3 True zero copy; 2.4 Synchronisation; 2.5 Tripwires for synchronisation; 3 Low Level Software; 3.1 Endpoints; 3.2 Apertures; 3.3 Tripwires; 3.4 DMA; 3.5 Multiple endpoints; 4 Programming the NIC; 4.1 Simple message transfer; 4.2 A distributed queue; 4.3 Discussion; 5 Raw Performance; 5.1 System characterisation; 5.2 Shared memory; 5.3 An event-based server; 5.4 Round-trip time; 5.5 Bandwidth; 6 Supporting Applications; 6.1 IP; 6.2 NFS; 6.3 MPl; 7 Related Work; 8 Conclusions; 9 Future Work; Acknowledgements; References

SIMULATION OF SELF-SIMILAR TRAFFIC AND A TCP TRAFFIC SIMULATOR1 Introduction; 2 Problems Put Forward; 3 Scheme of Simulation; 4 Parameter Estimations of Impulse Functions; 5 A Type of Impulse Functions for Simulation of TCP Traces; 5.1 Simulation of TCP Trace; 5.2 Filters Design based on Nonlinear Least Squares Fitting; 6 Conclusions; 7 Acknowledgements; References; ON THE REARRANGEABILITY OF SHUFFLE-EXCHANGE NETWORKS; 1 Introduction; 2 Preliminaries; 3 Main Results; 4 Discussions; Acknowledgments; References; PREDICTABILITY OF MESSAGE TRANSFER IN CSMA-NETWORKS; 1 Introduction

2 Controlling the access to the network

Notes:

"The fourth IEEE International Conference on Algorithms and Architecture for Parallel Processing (ICA3PP2000)"--Pref.

Includes bibliographical references and index.

ISBN:

9789812792037

9812792031