Parallel Cluster Computing with IEEE1394-1995

Authors:
László Böszörményi;Günther Hölzl;Emanuel Pirker
Affiliations:
-;-;-
Venue:
ParNum '99 Proceedings of the 4th International ACPC Conference Including Special Tracks on Parallel Numerics and Parallel Computing in Image Processing, Video Processing, and Multimedia: Parallel Computation
Year:
1999

Citing 7
Cited 0

Discrete-time signal processing

Discrete-time signal processing
Design, implementation, and evaluation of a software-based real-time Ethernet protocol

SIGCOMM '95 Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Reliable broadcast protocols

ACM Transactions on Computer Systems (TOCS)
Interconnection Networks for Multiprocessors and Multicomputers: Theory and Practice

Interconnection Networks for Multiprocessors and Multicomputers: Theory and Practice
Myrinet: A Gigabit-per-Second Local Area Network

IEEE Micro
A Comparison of Three High Speed Networks for Parallel Cluster Computing

CANPC '97 Proceedings of the First International Workshop on Communication and Architectural Support for Network-Based Parallel Computing
Multicast Communication on Network Computers

IEEE Software

Quantified Score

Hi-index	0.00

Visualization

Abstract

Diverging demands on computer networks, such as high bandwidth, guaranteed quality of service and low latency lead to growing heterogeneity. IEEE1394-1995 is a standardized low-cost highperformance serial-bus-system with both isochronous and asynchronous operation. It might be an interesting candidate for all-round local and system area networks, providing a good compromise in fulfilling the above demands for low costs. Beside providing some technical background we show the possibilities and advantages of building parallel clusters on top of IEEE1394-1995. The main advantage is that minimal speed-up can be guaranteed, as shown on the basis of the parallel implementation of discrete Fourier transformation.