eugene@orville.nas.nasa.gov (Eugene Miya) (10/12/88)
The following are some interesting new books by Elsevier (a series).
I recommend them for ORGANIZATIONS, their cost is a bit stiff for
individuals (but who knows maybe you want your own copy? ;-).
I work with one of the editors. I have also included a couple of
other books and some other interesting articles and TRs.
--eugene
%A J. J. Dongarra, ed.
%Z Argonne National Lab, IL
%T Experimental Parallel Computing Architectures
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%K book, text,
%X There are a lot of previously published papers in this book.
Either not many people have much to do, or it's a good thing to
put them all together.
%A David J. Kuck
%A Edward S. Davidson
%A Duncan H. Lawrie
%A Ahmed H. Sameh
%Z CSRD, U. Ill.
%T Parallel Supercomputing Today and the Cedar Approach
%B Experimental Parallel Computing Architectures
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 1-23
%K paracompiler, Alliant FX/8, doacross,
%X A paper with same title but shorter appears in "Science,"
vol. 231, no. 4741, February 28, 1986, pages 967-974, survey
and introduction to the project (as opposed to the whole field).
Lacks discussion about algorithms.
%A Allan Gottlieb
%Z Courant Institute of Mathematical Sciences, Computer Science Dept.,
New York University
%T An Overview of the NYU Ultracomputer Project
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 25-95
%r ULTRACOMPUTER NOTE #32
%r ULTRACOMPUTER NOTE #100
%K grecommended, paracomputer, WRAM, fetch-op, parbegin-parend,
barriers, scheduling, Omega network, Unix, fetch-add,
%A Geoffrey C. Fox
%Z Physics, Dept., Caltech
%T Questions and Unexpected Answers in Concurrent Computation
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 97-121
%K hypercubes, Caltech concurrent computation program, JPL,
Mark I, Mark II, Mark III, interesting! grecommended
%X Has an interesting time table of development. This chapter is different.
Some of the questions are:
can quantum field theory be solved numerically on a concurrent computer?
why do scientists need large amounts of computer time? (Good Q!)
can hypercubes be generally used in science and engineering? (good Q!)
how did the use like the hypercube and how do we help them? (good Q, but
answer is a waffle.)
what is the right design for a hypercube? (good Q!)
are we computing science and/or scientifically studying computers? (good Q!)
issues in concurrent algorithms. what are the appropriate hardware
and software environments and architectures? (Fox tries to talk to Cray
users without success...).
%A G. F. Pfister
%A W. C. Brantley
%A D. A. George
%A S. L. Harvey
%A W. J. Kleinfelder
%A K. P. McAuliffe
%A E. A. Melton
%A V. A. Norton
%A J. Weiss
%T An Introduction to the IBM Research Parallel Processor Prototype (RP3)
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 123-140
%K The IBM Research Parallel Processor, RP3,
MIMD, omega net,
grecommended,
%X Also see Ultracomputer work. 1.3 GIPS, 800 MFLOPS, 1-2 Gbytes main storage,
192 MB/sec I/O, 13 Gbytes/sec interprocessor rate, 1 GIP sustained rate.
They looked at Ultracomputer and Cosmic Cube. Lots of close work with
Ultra: using 4.2BSD as OS and C as parallel language.
Net is bipolar technology. Processor element: 32-bit micro, 2-4MB main,
32 KB cache, 801 RISC type CPU, NMOSFET technology. Network will have
fetch-op capability of synchronization. VM/EPEX is a software emulator
for the modified 4.2BSD.
Reproduced in "Parallel Computing: Theory and Comparisons,"
by G. Jack Lipovski and Miroslaw Malek,
Wiley-Interscience, New York, 1987, pp. 270-280, appendix C.
%X Looks a lot like the paper in the Proceedings of the 1985
International Conference on Parallel Processing,
IEEE, University Park, Pennsylvania, August 1985, pages 764-771
with the title turned around or IBM TR RC 11060 (#49693), March 1985,
IBM T. J. Watson Research Center, Yorktown Heights, NY.
%A Enrico Clementi
%A John Detrich
%Z IBM
%T Largel Scale Parallel Computations on a Loosely Coupled Array of Processors
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 141-176
%K lCAP, IBM 43xx, 3090s, FPS array processors, precompiler,
%A John Gurd
%A Chris Kirkham
%A Wim Bohm
%Z U. Manchester
%T The Manchester Dataflow Computing System
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 177-219
%K SISAL, reduction, ring, matching store, throttling,
%A Alexandru Nicolau
%Z Cornell U.
%T The Cornell Parallel Supercomputing Effort
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 221-253
%K arrays processors, GIBBs, Ken Wilson,
%X IBM oriented. Boring.
%A John Beetem
%A Monty Denneau
%A Don Weingarten
%Z IBM TJW, Yorktown Heights
%T The GF11 Supercomputer
%E J. J. Dongarra
%S Special Topics in Supercomputing
%V 1
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 255-298
%K Quantum chromodynamics,
%X 576 processors, modified SIMD, 2 MB memory per processor at 20 MFLOPS.
Memphis Switch, 50 ns. 1,125 Macho Bytes, 11,520 Macho FLOPS.
See also the Proceedings of the 12th International Symposium on
Computer Architecture, IEEE, June 1985, Boston, MA, pages 108-115
or see the IBM TR RC 10852 from TJW.
%A Willi Schonauer
%Z Univ. of Karlsruhe, FRG
%T Scientific Computing on Vector Computers
%S Special Topics in Supercomputing
%V 2
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%K book, text, Hockney's $n sub {1 over 2}$,
Fortran, vectorization, matrix operations, algorithms,
Fujitsu VP, CRAY-2, IBM VF, Convex C-1, FIDISOL, ETA-10,
chaining, indirect addressing and indexing, direct vectorization,
linear first order recurrence, recursive doubling, cyclic reduction,
unrolling loops, Gaussian elimination, Gauss-Jordan, LU decomposition,
Householder algorithm, Crout algorithm, pivoting,
tridiagonal linear systems, pentadiagonal, boundary value problems,
Jacobi, successive overrelaxation (SOR), conjugate gradient (CG),
finite difference methods (FDM), finite element methods (FEM),
MJOR, Gauss-Seidel, colored SOR, sparse matrix (SM), multigrid (MG),
fast Fourier Transform (FFT),
%X Why I like vector computers.
%X Interesting comment on supercomputers. An interesting book on
vector machines: discusses algorithms for vectorization (as opposed to
numerical algorithms [that, too]). It's written in a casual but
strong technical style which might grate on some nerves.
%A H. J. Te\ Riele
%A Th. J. Dekker
%A H. A. van\ der\ Vorst, ed.
%T Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%$ 88
%K Book, text,
%X This was a series of monthly colloquia 1985-1986. (CWI) Centre for
Mathematics and Computer Science, Amsterdam. It has a bit or
parochialism and volume 3 isn't as good as volume 2.
%A Ad H. H. Emmen
%Z SARA, Amsterdam
%T A Survey of Vector and Parallel Processors
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 1-45
%K general aspects, Delft Intel iPSC, MAP-6420, Alliant FX/8,
FPS T-series, Tesseract, IBM 3090 VF, Cray-[12], Cyber 205, ETA-10,
X-MP, Fujitsu FACOM VP-200, NEC SX,
%X Another one (survey), sigh! Starts with a classification of taxonomies
(Flynn, Schwartz, and Hockney, note it left out the German one 1977:
Handler intentional?). Note: no MPP, no Connection Machine.
The keyword list has the major machines, others are listed in tables
but not covered in detail. A little horn tooting.
%A D. Hilhorst
%A H. D. A. Bram
%A H. J. Sips
%T Some Methods for the Automatic Parallelization of Programs;
the Hyperplane method and its Application on the Delft Parallel Processor
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 47-73
%K general aspects, ODEs, PDEs, FORTRAN, code restructuring,
redistribution, DPP 84,
%X Chapter mentions "Bernstein's condition" without citing it.
%A Erik H. D'Hollander
%Z State Univ. of Ghent
%T Computer Aided Dataflow Analysis for the Conversion of Sequential Programs
into Parallel Form
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 75-102
%K general aspects, data driven, demand driven, computation,
%A A. M. Kobos
%A R. E. VanKooten
%A M. A. Walker
%Z Myrias, Alberta
%T The Myrias parallel processor computer system
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 103-118
%K general aspects, par DO, G virtual machine, G language,
%A Jan van\ Kats
%A Aad van\ der\ Steen
%A Rossend LLurba
%T Result of a benchmark test on a Siemens VP-200 vector processor with
comparisons to other supercomputers
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 117-144
%K general aspects, Fujitsu VP-200,
vector optimization control lines (VOCL), compiler directives,
FORTUNE, VECTUNE, editors, interactive vectorizers,
%A D. T. Winter
%Z Centre for Mathematics and Computer Science, Amsterdam
%T Efficient use of memory and Input/Output
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 145-150
%K general aspect, virtual memory, matrices, paging,
two-dimension arrays,
%A D. T. Winter
%Z Centre for Mathematics and Computer Science, Amsterdam
%T Scalar optimization
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 151-154
%K general aspect, hints, Cyber 205,
%X Has some architecture independent hints.
%A S. W. Brok
%Z Delft University, Netherlands
%T Simulation of gas pulsations in pipe systems on the MIMD-structured
Delft Parallel Processor (DPP/84)
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 155-165
%K implementations, hyperbolic partial differential equations,
%A Walter Hoffmann
%Z Univ. of Amsterdam
%T Solving linear systems by direct methods related to Gaussian
elimination
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 167-179
%K implementations, Crout's factorization, Doolittle's factorization,
%X Introduces a notation on the structure of elements, rows and
columns and their corresponding algorithms.
%A W. M. Lioen
%Z Centre for Mathematics and Computer Science, Amsterdam
%T Multigrid methods for elliptic PDEs
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 181-198
%K implementations, NUMVEC numerical software for vector computers,
scatter-gather,
%A W. M. Lioen
%A M. Louter-Nool
%A H. J. J. te\ Riele
%Z Centre for Mathematics and Computer Science, Amsterdam
%T Optimization of the real Level 2 BLAS on the Cyber 205
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 199-212
%r CWI. Department of Numerical Mathematics ; R 8715
%M f 4,--
%K implementations, matrix vector, multiplication,
solution of triangular systems
%X Covers some of the unique Q8 subroutine calls of the Cyber.
%A Walter Ravenek
%Z Scheikundig Laboratorium der Vrije Univ., Netherlands
%T Vectorization of Quantum Chemical Programs: the HFS-LCAO package
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 213-227
%K implementations, Hartree-Fock-Slater method,
%A J. Schlichting
%Z Centre for Mathematics and Computer Science, Amsterdam
%T Double Precision BLAS: Implementation design for Double Precision
Basic Linear Algebra Subprograms on the Control Data Cyber 205
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 229-249
%K implementations, FTN200,
%A F. W. Wubs
%Z Centre for Mathematics and Computer Science, Amsterdam
%T An Explicit Shallow-Water Equations Solver for use on the CYBER 205
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 251-277
%K implementations, AMS (MOS) classification: primary, secondary,
vector computer, hyperbolic equations, SWE, Runge-Kutta,
%A O. Axelsson
%A V. Eijkhout
%Z Univ. Nijmegen, Netherlands
%T Robust Vectorizable Preconditioners for Three-Dimensional Elliptic
Difference Equations with Anisotropy
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 279-306
%K numerical algorithms, factorization, differencing, PCG,
%A A. Bojanczyk
%A R. Brent
%A P. van\ Dooren
%A F. de\ Hoog
%T Downdating the Cholesky factorization
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 307-323
%K numerical algorithms, LINPACK, recursive, modified recursive,
%A E. J. H. Kerckhoffs
%Z Delft University, Netherlands
%T Multiprocessor algorithms for ODEs
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 325-345
%K numerical algorithms, MIMD, Nievergelt's method, Sameh and Brent,
Luck and Muraoka, recurrence equations, Yura, equation segmentation,
%A Dianne P. O'Leary
%Z Univ. Maryland
%T Fine and Medium Grained Parallel Algorithms for Matrix QR
Factorization (Extended Abstract)
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 347-349
%K numerical algorithms,
%A M. H. C. Paardekooper
%Z Tilburg, Univ., Netherlands
%T Sameh's parallel eigenvalue algorithm revisited
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 351-371
%K numerical algorithms, shears,
%A Hans M. Amman
%Z Univ. of Amsterdam, Netherlands
%T Applying supercomputers for optimal control experiments in
economics
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 373-378
%K applications, measurements,
%A J. B. Vos
%Z Delft University, Netherlands (now ETH)
%T The Calculation of Chemical Reacting Turbulent Boundary Layers
using the Cray-1 Supercomputer
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 379-403
%K applications, solid fuel calculations, combustion, kinetics,
%A G. M. van\ Waveren
%T Application of Sparse Vector Techniques on a Molecular Dynamics
Program
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 405-428
%K applications, Cyber 205, compress-expand,
%A Wouter Zijl
%A Marek Nawalany
%A Frank Pasveer
%T Numerical Simulation of Fluid Flow in Porous Media using the
Cyber 205 and the Delft Parallel Processor
%E H. J. Te\ Riele
%E Th. J. Dekker
%E H. A. van\ der\ Vorst
%B Algorithms and Applications on Vector and Parallel Computers
%S Special Topics in Supercomputing
%V 3
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1987
%P 429-453
%K applications, reservoir simulation, hydrology, sparse matrix,
%A Joanne L. Martin, ed.
%Z IBM TJW
%T Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%K book, text,
%X This book is composed from end-user perspectives of performance
rather than manufacturer, or performance evaluation community.
It is a little something of a survey and a political statement
[See Denning paper.]
%A Joanne L. Martin
%Z IBM TJW
%T Supercomputer Performance Evaluation: The Comparative Analysis of
High-Speed Architectures Against Their Applications
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 3-20
%K benchmarking, analytic queueing models,
%X Introduction. Overview.
%A Leah H. Jamieson
%Z Purdue, EE Dept.
%T Using Algorithm Characteristics to Evaluate Parallel Architectures
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 21-49
%K benchmarking, virtual algorithm, mapping, matching,
image processing orientation, complexity,
%X Important work, but just scratching the surface. Better to see her
book with Gannon and Douglass "The Characteristics of Parallel Algorithms,"
MIT Press, 1987.
%A James R. B. Davis
%Z Kuck Assoc.
%T Issues in Compiler Performance
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 51-68
%K scalar optimization, statements, blocks, flow, interprocedure,
vectorization, data dependence, anti-dependence,
loop distribution and interchange, reduction, expansion, index
variables,
induction variables, concurrency, code generation,
%A Elizabeth Williams
%Z LANL (now SRC)
%T The Effects of Operating Systems on Supercomputer Performance
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 69-81
%K CRAY X-MP, CRAY-2, CTSS, UNICOS, environment, overhead, wait time,
I/O, memory, spin wait, measurement, benchmarking,
%A Donald A. Calahan
%A David H. Bailey
%T Measurement and Analysis of Memory Conflicts on Vector
Multiprocessors
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 83-106
%K CRAY-2, CRAY X-MP, UNICOS, environment, delays, sensitivity,
dynamic and static access,
%A Kyle A. Gallivan
%A William Jalby
%A Allen D. Malony
%A Pen-Chung Yew
%Z CSRD, U. Ill.
%T Performance Analysis on the CEDAR System
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 109-142
%K measurement and metrics,
%r CSRD Rpt. No. 680
%i University of Illinois
%c Urbana-Champaign, IL
%d September 1987
%X Slight differences in the report and the Chapter.
%A Frank McMahon
%Z LLNL
%T The Livermore FORTRAN Kernels Test of the Numerical Performance
Range
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 143-186
%K measurement and metrics, Livermore loops,
%X See the technical report UCRL-53745, Dec. 1986 for more complete details.
Same author and title at LLNL.
%A Kimming So
%A Frederica Darema
%A David A. George
%A V. Alan Norton
%A Gregory F. Pfister
%Z IBM TJW
%T PSIMUL - A System for Parallel Simulation of Parallel Systems
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 187-213
%K measurement and metrics,
%K simulation of computers, parallel simulation, multiprogramming,
multiprocessing, parallel programs, memory access patterns, cache,
VM/EPEX, RP3, SEMUL,
%X See also IBM TWJ TR RC 11674 January 1986 by the same authors
possibly an earlier version.
%A Roger W. Hockney
%Z Reading Univ.
%T Problem Related Performance Parameters for Supercomputers
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 215-235
%K measurement and metrics, hardware, SIMD, MIMD, vector performance,
$r sub inf, n sub {1 over 2 }$ benchmark, MIPS, MFLOPS,
$r hat sub inf, n hat sub {1 over 2 }, f sub {1 over 2}$ benchmark,
algorithmic and phase analysis,
%A James C. Browne
%A Ashok K. Adiga
%Z U. Texas, Austin
%T Graph Structured Performance Models
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 239-281
%K methods, models, and directions,
[standard, [generalized] timed] Petri nets (PN, SPN, GTPN, TPN),
UCLA Graph Model of Behavior (GMB), Parallel Computation Model (PCM),
hierarchical graph model, H-graph, dependency graph model,
%A Maria Calzarossa
%A Giuseppe Serazzi
%T Workload Characterization for Supercomputers
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 283-315
%K methods, models, and directions, real (live) and artificial
workloads,
executable and nonexecutable, static and dynamic models,
performance metrics, tools, Monit, Amdahl SMF, Unisys PEER,
workload construction, benchmarking,
%X This paper has a Kiviat diagram without labelling it so. The
section on construction is weak.
%A Eugene N. Miya
%Z NASA Ames Research Center
%T Empirical Computer Performance Measurement
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 317-340
%K methods, models, and directions, static and dynamic measurement,
observation, performance artifacts, experiment design,
experiment invalidity and interaction, minimally equivalent benchmark,
open benchmarking,
%X This paper is generally weak, and the ideas were not well formed
when I wrote them. While the basic intent is the same, I would
rewrite the ideas. The area needs more rigour.
%A Dieter Muller-Wichards
%Z German Aerospace Research Establishment (DFVLR)
%T Performance Estimates for Applications: An Algebraic Framework
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 341-402
%i IBM T.J. Watson Research Center
%c Yorktown Heights, NY
%r RC 12391 (#55621)
%d December 1986
%K methods, models, and directions, static measurement,
performance algebra, semigroup,
%K Performance modelling, ordered semigroups, parallel - sequential
expressions, homomorphisms, arithmetic operations, data movements,
delays, sensitivity analysis
%X Some conventional queueing network types have expressed some dismay
about this work.
%A Peter J. Denning
%A George B. Adams, III
%Z RIACS
%T Research Questions for Performance Analysis of Supercomputers
%E Joanne L. Martin
%B Performance Evaluation of Supercomputers
%S Special Topics in Supercomputing
%V 4
%I Elsevier Science Publishers B.V. (North-Holland)
%C Amsterdam
%D 1988
%P 403-419
%K methods, models, and directions,
%A H.X. Lin
%A H.J. Sips
%T Parallel Vector Reduction Algorithms and Architectures
%J Journal of Parallel and Distributed Computing
%V 5
%N 2
%D April 1988
%P 103-130
%A Charles V. Stewart
%A Charles R. Dyer
%T Scheduling Algorithms for PIPE (Pipelined Image-Processing Engine)
%J Journal of Parallel and Distributed Computing
%V 5
%N 2
%D April 1988
%P 131-153
%A Mi Lu
%A Peter Varman
%T Optimal Algorithms for Rectangle Problems on a Mesh-Connected
Computer
%J Journal of Parallel and Distributed Computing
%V 5
%N 2
%D April 1988
%P 154-171
%A Franklin T. Luk
%A Haesun Park
%T RESEARCH NOTES, An Analysis of Algorithm-Based Fault Tolerance
Techniques
%J Journal of Parallel and Distributed Computing
%V 5
%N 2
%D April 1988
%P 172-184
%A S.C. Kothari
%A G.M. Prabhu
%A Robert Roberts
%T A Multipath Network with Cross Links
%J Journal of Parallel and Distributed Computing
%V 5
%N 2
%D April 1988
%P 185-193
%A Nicola Santoro
%A Michael Scheutzow
%A Jeffrey B. Sidney
%T On the Expected Complexity of Distributed Selection
%J Journal of Parallel and Distributed Computing
%V 5
%N 2
%D April 1988
%K 194-203
%A Christopher Eoyang
%T The Hitachi S-820/80 Supercomputer
%I ISR (Institute for Supercomputing Research)
%J Vector Register
%V 1
%N 4
%D March 1988
%A J. R. Kirkland
%A J. H. Poore, ed.
%T Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%I Praeger
%C New York
%D 1987
%K Book, Text,
%X Not impressed. Largely a socio-economic, political statement.
%A J. R. Kirkland
%A J. H. Poore
%T Initiatives for Policies and Programs
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 3-8
%A George Kozmetsky
%T Supercomputers and National Policy: Maintaining U.S. Preeminence in
an Emerging Industry
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 9-42
%A J. H. Poore
%T Perspective on a National Science Network
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 43-49
%A Lloyd M. Thorndyke
%T Supercomputer Systems Markets
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 51-56
%A F. Brett Berlin
%T A Policy Opportunity
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 57-76
%A R. David Lowry
%T The Parallel Processing Revolution
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 77-84
%A Wayne McIntyre
%T Amdahl Vector Processors
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 85-108
%A W. Daniel Hillis
%T Why Parallel Processing Is Inevitable
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 119-122
%A John Killeen
%T The Role of Supercomputers in Energy Research Programs
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 123-128
%A Randolph Graves
%T Supercomputers at NASA
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 129-136
%A Glenn Ingram
%T National Bureau of Standards: Outreach to the Supercomputer
Community
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 137-143
%A Ken Wilson
%T The Cornell Approach to Training and Research
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 145-150
%A Larry Smarr
%T The Illinois System
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 151-156
%A Sidney Karin
%T The San Diego Supercomputer Center
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 157-164
%A Robert Johnson
%A Joseph Lannutti
%T The Collaborative Venture at Florida State University
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 165-172
%A Walter McRae
%T The Independent Center at the University of Georgia
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P173-182
%A George Dodd
%T Productivity, Cost, and Quality: Supercomputers at General Motors
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 183-188
%A Henry Zanardelli
%T The Emerging Need for Supercomputer-Trained Engineers at Ford
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 189-192
%A Edwin B. Neitzel
%T The Energy Industry: Supercomputer Technology at ARCO
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 193-202
%A David Pensak
%T The Chemical Industry: Training Supercomputer Scentists at Du Pont
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 203-208
%A Peter Zaphyr
%T The Westinghouse Network
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 209-216
%A James C. Browne
%A John Feo
%A Patricia Roe
%T Bibliography of Literature on Supercomputers
%B Supercomputers: A Key to U.S. Scientific, Technological, and
Industrial Preeminence
%E J. R. Kirkland
%E J. H. Poore
%I Praeger
%C New York
%D 1987
%P 217-228
%X Oh, yeah?! 8-)
%A J. L. Potter
%T Data Structures for Associative Supercomputers
%R SRC-TR-87-003
%I Supercomputing Research Center
%C Lanham, MD
%D January 1987
%A C. Henery Chu
%A Edward J. Delp
%A Leah H. Jamieson
%A Howard Jay Siegel
%A Frank J. Weil
%A Andrew B. Whinston
%T A Model for an Intelligent Operating System for Executing Tasks on
a Reconfigurable Parallel Architecture
%R SRC-TR-87-007
%I Supercomputing Research Center
%C Lanham, MD
%D November 1987
%A Sans Pareil
%T Isomorphic Computers Inc.
%J International Journal of Parallel Programming
%V 16
%N 2
%D 1987
%P 179-182
%K humor, satire, speed up, performance measurement, comparison conditions,
%X Similar to "How to Lie with Statistics" but shorter. Good reading.
%A R. H. Perrott
%T Parallel Programming
%S International Computer Science Series
%I Addison-Wesley
%C Menlo Park
%D 1987
%$ 27
%K Book, text, Pascal Plus, communication, Modula-2, Ada, Occam,
mutual exclusion, synchronization, CFT, CFD, Cyber Fortran, FTN,
distributed array processor, DAP, mesh, ACTUS, Cray, dataflow,
message passing,
%X An earlier edition of this book was published in 1985.
Book does not cover hypercubes.