Heterogeneous High Performance Computer

Authors:
Virginia W. Ross
Affiliations:
US Air Force Research Laboratory/Information Directorate (AFRL/IF), Rome, NY
Venue:
DOD_UGC '05 Proceedings of the 2005 Users Group Conference on 2005 Users Group Conference
Year:
2005

Citing 0
Cited 3

Parallel backprojection: a case study in high-performance reconfigurable computing

EURASIP Journal on Embedded Systems - FPGA supercomputing platforms, architectures, and techniques for accelerating computationally complex algorithms
Content-based image retrieval algorithm acceleration in a low-cost reconfigurable FPGA cluster

Journal of Systems Architecture: the EUROMICRO Journal
Pub/Sub on stream: a multi-core based message broker with QoS support

Proceedings of the 6th ACM International Conference on Distributed Event-Based Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper discusses the proven concept of marrying the computing capabilities of a cluster computer with the reconfigurability of adaptive computing systems into a 48 node heterogeneous high performance computer (HHPC) with 34 TOPS peak performance. This system has shown over two years of success supporting the Signal and Image Processing (SIP), Forces Modeling and Simulation (FMS), Computational Electromagnetics and Acoustics (CEA), and Integrated Modeling and Test Environments (IMT) communities. The HHPC supports both local and remote users. Highlights of the research efforts performed on this system, and its advancement of reconfigurable computing, since it was introduced in 2003, are discussed.