Skip to main content
    Country/region [change]    Terms of use
    Home    Products    Services & solutions    Support & downloads    My account    

IBM Journal of Research and Development

Hybrid Computing Systems Volume 53, Number 5, 2009
RD 53-5
Difficult technological challenges facing microprocessor design are forcing a rethinking of how best to utilize the increasing number of transistors provided by Moore's Law. One promising approach is to tailor system characteristics to specific applications. These new "hybrid" system designs combine multiple types of processor cores on a single chip to reduce system power and chip area while improving performance. The twelve papers in this issue describe hybrid system challenges and solutions from component design and optimization to system architectures, design tools, programming models, and applications. This issue explores hardware and software design for hybrid systems, ranging from the first supercomputer to achieve a performance of over one petaflops to medical image processing applications.
Table of Contents  
Preface M. Gschwind and M. Perrone, Guest Editors  
Breaking the petaflops barrier D. Grice, H. Brandt, C. Wright, P. McCarthy, A. Emerich, T. Schimke, C. Archer, J. Carey, P. Sanders, J. A. Fritzjunker, S. Lewis, and P. Germann Paper 1
directCell: Hybrid systems with tightly coupled accelerators H. Penner, U. Bacher, J. Kunigk, C. Rund, and H. J. Schick Paper 2
IBM BladeCenter QS22: Design, performance, and utilization in hybrid computing systems J.-S. Vogt, R. Land, H. Boettiger, Z. Krnjajic, and H. Baier Paper 3
Integrated execution: A programming model for accelerators M. Gschwind Paper 4
Applying Amdahl’s Other Law to the data center D. Cohen, F. Petrini, M. D. Day, M. Ben-Yehuda, S. W. Hunter, and U. Cummings Paper 5
Software architecture and system validation of an open, unified model for accelerated multicore computing C. H. Crawford, D. J. Burdick, J. N. Dale, E. F. Ford, R. A. Mikosh, A. Nobles, and V. To Paper 6
Cell Broadband Engine processor performance optimization: Tracing tools implementation and use M. Biberstein, S. Dori-Hacohen, Y. Harel, A. Heilper, B. Mendelson, U. Shvadron, E. Treister, J. Turek, and M. S. Chang Paper 7
The reverse-acceleration model for programming petascale hybrid systems S. Pakin, M. Lang, and D. J. Kerbyson Paper 8
Programming the Linpack benchmark for Roadrunner M. Kistler, J. Gunnels, D. Brokenshire, and B. Benton Paper 9
MapReduce for the Cell Broadband Engine Architecture M. de Kruijf and K. Sankaralingam Paper 10
A parallel computing approach for tracking of neuronal fibers J. Roth, H. Schiffbauer, and P. Váterlein Paper 11
Accelerating 3D nonrigid registration using the Cell Broadband Engine processor J. Rohrer and L. Gong Paper 12

    About IBMPrivacyContact