Janusz S. Kowalik

  • Parallel MIMD Computation

    Parallel MIMD Computation

    HEP Supercomputer and Its Applications

    Janusz S. Kowalik

    Fifteen original contributions from experts in high-speed computation on multi-processor architectures, concurrent programming and parallel algorithms.

    Experts in high-speed computation agree that the rapidly growing demand for more powerful computers can only be met by a radical change in computer architecture, a change from a single serial processor to an aggregation of many processors working in parallel. At present, our knowledge about multi-processor architectures, concurrent programming or parallel algorithms is very limited. This book discusses all three subjects in relation to the HEP supercomputer that can handle multiple instruction streams and multiple data streams (MIMD). The HEP multiprocessor is an innovative general purpose computer, easy to use by anybody familiar with FORTRAN. Following a preface by the editor, the book's fifteen original contributions are divided into four sections: The HEP Architecture and Systems Software; The HEP Performance; Programming and Languages; and Applications of the HEP Computer. An appendix describes the use of monitors in FORTRAN, providing a tutorial on the barrier, self-scheduling DO loop, and Askfor monitors.

    J. S. Kowalik, who has contributed a chapter with S. P. Kumar on "Parallel Algorithms for Recurrence and Tridiagonal Linear Equations," is a manager in Boeing Computer Services' Artificial Intelligence Center in Seattle.MIMD Computation is included in the Scientific Computation Series, edited by Dennis Cannon.

    • Hardcover
    • Paperback $55.00

Contributor

  • Data-Parallel Programming on MIMD Computers

    Data-Parallel Programming on MIMD Computers

    Philip J. Hatcher and Michael J. Quinn

    Data-Parallel Programming demonstrates that architecture-independent parallel programming is possible by describing in detail how programs written in a high-level SIMD programming language may be compiled and efficiently executed-on both shared-memory multiprocessors and distributed-memory multicomputers.

    MIMD computers are notoriously difficult to program. Data-Parallel Programming demonstrates that architecture-independent parallel programming is possible by describing in detail how programs written in a high-level SIMD programming language may be compiled and efficiently executed-on both shared-memory multiprocessors and distributed-memory multicomputers. The authors provide enough data so that the reader can decide the feasibility of architecture-independent programming in a data-parallel language. For each benchmark program they give the source code listing, absolute execution time on both a multiprocessor and a multicomputer, and a speedup relative to a sequential program. And they often present multiple solutions to the same problem, to better illustrate the strengths and weaknesses of these compilers. The language presented is Dataparallel C, a variant of the original C* language developed by Thinking Machines Corporation for its Connection Machine processor array. Separate chapters describe the compilation of Dataparallel C programs for execution on the Sequent multiprocessor and the Intel and nCUBE hypercubes, respectively. The authors document the performance of these compilers on a variety of benchmark programs and present several case studies.

    ContentsIntroduction • Dataparallel C Programming Language Description • Design of a Multicomputer Dataparallel C Compiler • Design of a Multiprocessor Dataparallel C Compiler • Writing Efficient Programs • Benchmarking the Compilers • Case Studies • Conclusions

    • Hardcover $9.75