article

Free access

Data broadcasting in linearly scheduled array processors

Authors:

J. A. B. Fortes,

D. I. MoldovanAuthors Info & Claims

ACM SIGARCH Computer Architecture News, Volume 12, Issue 3

Pages 224 - 231

https://doi.org/10.1145/773453.808186

Published: 01 January 1984 Publication History

Abstract

A major problem in executing algorithms in array processors is the implementation of broadcasts without unnecessary speed-up factor degradation. We discuss when and how broadcasts can be eliminated or reduced to easily implementable sequences of reduced local broadcasts. Algorithms are modelled as a structured set of indexed computations which operate on variables associated with a referencing or indexing function. The discussion is restricted to variables with linear indexing functions and to algorithms linearly scheduled for execution in array processors. Linear indexing functions are represented as affine matricial functions of the index set of the algorithm. The linear part of such representation is a coefficient matrix denoted the indexing matrix. Linear schedules are defined as linear time-space allocation functions mapping the computations of an algorithm into time and processors. We discuss necessary and sufficient conditions for the occurrence of broadcasts in a linearly scheduled algorithm. Necessary and sufficient conditions and constructive criteria are given for selecting linear schedules for which all broadcasts are eliminated or reduced to sequences of small local broadcasts.

References

[1]

Barnes et al., "The ILLIAC-VI computer," IEEE Trans. on Computers, C-17, Aug. 1968, pp. 746-757.

[2]

K. E. Batcher, "Design of a massively parallel processor," IEEE Trans. on Computers, C-29, pp. 836-840.

Digital Library

[3]

J. A. B. Fortes, "Algorithm transformations for parallel processing and VLSI architecture design," Ph.D. Thesis, Dept. of EE-Systems, University of Southern California, December 1983.

Digital Library

[4]

J. A. B. Fortes, D. I. Moldovan, "Parallelism detection and algorithm transformation techniques useful for VLSI designs," Tech. Report. PPP83-1, Dept. of EE-Systems, Univ. of Southern Calif., 1983.

[5]

K. Hwang, Y-H Chen, "Partitioned matrix algorithms for VLSI arithmetic systems," IEEE Trans. on Computers, C-31, No. 12, December 1982, pp. 1215-1224.

[6]

D. J. Kuck, "The structure of computers and computations," Vol. 1, John Wiley & Sons, Inc., 1978.

Digital Library

[7]

D. J. Kuck, et al., "Dependence graphs and compiler optimizations," in Proc. 8th ACM Symp. Principles of Programming Languages, Jan. 1981, pp. 207-218.

Digital Library

[8]

R. H. Kuhn, "Optimization and interconnection complexity for parallel processors, single stage networks and decision trees," Ph.D. Thesis, Dept. of Computer Science, Report 80-1009, Univ. of Illinois, Urbana-Champaign, Ill., Feb. 1980.

Digital Library

[9]

H.T. Kung, "Let's design algorithms for VLSI systems," in Proc. of Caltech Conf. on VLSI, California Institute of Technology, Jan, 1979.

[10]

S. Y. Kung, "Wavefront array processor: language, architecture and applications," IEEE Trans. on Computers, Vol. C-31, No. 11, Nov. 1982, pp. 1054-1066.

[11]

L. Lamport, "The parallel execution of DO-loops," Comm. of ACM, Vol. 17, No. 2, February 1976, pp. 83-93.

Digital Library

[12]

C. E. Leiserson, J. B. Saxe, "Optimizing synchronous systems," in Proc. 22nd Annual Symposium on Foundations of Computer Science, IEEE Computer Society, Oct. 1981, pp. 23-34.

[13]

D. I. Moldovan, "On the analysis and synthesis of VLSI algorithms," IEEE Trans. on Computers, Vol. C-31, No. 11, Nov. 1982.

[14]

D. I. Moldovan, "On the design of algorithms for VLSI systolic arrays," Proc. of IEEE, Vol. 71, No. 1, pp. 113-120, Jan. 1983.

[15]

D. I. Moldovan, J. A. B. Fortes, "Partitioning algorithms for fixed size VLSI architectures," Tech. Report PPP83-5, Dept. of EE-Systems, University of Southern California, 1983.

[16]

D. Nassimi, S. Sahni, "Data Broadcasting in SIMD computers," IEEE Trans. on Comp., Vol. C-30, No. 2, Feb. 1982.

[17]

L. Snyder, "Introduction to the configurable highly parallel computer," Computer, Vol. 15, No. 1, Jan. 1982, pp. 47-64.

Digital Library

Cited By

Rajopadhye S(2010)Dependence Analysis and Parallelizing TransformationsThe Compiler Design Handbook10.1201/9781420040579.ch10Online publication date: 7-Mar-2010
https://doi.org/10.1201/9781420040579.ch10
Evans DGuŠev M(1997)Algorithm transformations for computational and data broadcastInternational Journal of Computer Mathematics10.1080/0020716970880459865:1-2(19-56)Online publication date: Jan-1997
https://doi.org/10.1080/00207169708804598
Shang WHodzic EChen Z(1996)On Uniformization of Affine Dependence AlgorithmsIEEE Transactions on Computers10.1109/12.50832145:7(827-840)Online publication date: 1-Jul-1996
https://dl.acm.org/doi/10.1109/12.508321
Show More Cited By

Recommendations

Data broadcasting in linearly scheduled array processors
ISCA '84: Proceedings of the 11th annual international symposium on Computer architecture

A major problem in executing algorithms in array processors is the implementation of broadcasts without unnecessary speed-up factor degradation. We discuss when and how broadcasts can be eliminated or reduced to easily implementable sequences of reduced ...
Register File Design Considerations in Dynamically Scheduled Processors
HPCA '96: Proceedings of the 2nd IEEE Symposium on High-Performance Computer Architecture

We have investigated the register file requirements of dynamically scheduled processors using register renaming and dispatch queues running the SPEC92 benchmarks. We looked at processors capable of issuing either four or eight instructions per cycle and ...
A Scalable Register File Architecture for Dynamically Scheduled Processors
PACT '96: Proceedings of the 1996 Conference on Parallel Architectures and Compilation Techniques

A major obstacle in designing dynamically scheduled processors is the size and port requirement of the register file. By using a multiple banked register file and performing dynamic result renaming, a scalable register file architecture can be ...

Comments

Information & Contributors

Information

Published In

cover image ACM SIGARCH Computer Architecture News

ACM SIGARCH Computer Architecture News Volume 12, Issue 3

June 1984

348 pages

ISSN:0163-5964

DOI:10.1145/773453

Issue’s Table of Contents

ISCA '84: Proceedings of the 11th annual international symposium on Computer architecture
January 1984
373 pages
ISBN:0818605383
DOI:10.1145/800015
Chairman:
Dharma P. Agrawal

Copyright © 1984 Authors.

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 01 January 1984

Published in SIGARCH Volume 12, Issue 3

Check for updates

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

34
Total Citations
View Citations
404
Total Downloads

Downloads (Last 12 months)65
Downloads (Last 6 weeks)13

Reflects downloads up to 12 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Rajopadhye S(2010)Dependence Analysis and Parallelizing TransformationsThe Compiler Design Handbook10.1201/9781420040579.ch10Online publication date: 7-Mar-2010
https://doi.org/10.1201/9781420040579.ch10
Evans DGuŠev M(1997)Algorithm transformations for computational and data broadcastInternational Journal of Computer Mathematics10.1080/0020716970880459865:1-2(19-56)Online publication date: Jan-1997
https://doi.org/10.1080/00207169708804598
Shang WHodzic EChen Z(1996)On Uniformization of Affine Dependence AlgorithmsIEEE Transactions on Computers10.1109/12.50832145:7(827-840)Online publication date: 1-Jul-1996
https://dl.acm.org/doi/10.1109/12.508321
MEGSON G(1993)THE DERIVATION OF UNIFORM RECURRENCE EQUATIONS FOR THE KNAPSACK PROBLEMParallel Algorithms and Applications10.1080/106371993089154361:2(127-140)Online publication date: Jan-1993
https://doi.org/10.1080/10637199308915436
Megson G(1993)Systolic partitioning algorithmsInformation Processing Letters10.1016/0020-0190(93)90189-G46:1(13-18)Online publication date: 29-Apr-1993
https://dl.acm.org/doi/10.1016/0020-0190%2893%2990189-G
Fernández ALiabería JNavarro JValero-García M(1991)Interleaving partitions of systolic algorithms for programming distributed memory multiprocessorsDistributed Memory Computing10.1007/BFb0032926(90-99)Online publication date: 1991
https://doi.org/10.1007/BFb0032926
Manolakos E(2010)A Modular VLSI Architecture for the Real- Time Estimation of Higher Order Moments and CumulantsHandbook of Parallel Computing and Statistics10.1201/9781420028683.ch12(375-398)Online publication date: 30-Jan-2010
https://doi.org/10.1201/9781420028683.ch12
Jebelean TSzakacs L(2005)Functional-Based Synthesis of Systolic Online MultipliersProceedings of the Seventh International Symposium on Symbolic and Numeric Algorithms for Scientific Computing10.1109/SYNASC.2005.36Online publication date: 25-Sep-2005
https://dl.acm.org/doi/10.1109/SYNASC.2005.36
Megson GRapanotti LChen X(2005)Automatic synthesis of parallel algorithmsSolving Combinatorial Optimization Problems in Parallel10.1007/BFb0027118(51-86)Online publication date: 17-Jun-2005
https://doi.org/10.1007/BFb0027118
Rajopadhye SFujimoto R(2005)Systolic array synthesis by static analysis of program dependenciesPARLE Parallel Architectures and Languages Europe10.1007/3-540-17943-7_135(295-310)Online publication date: 1-Jun-2005
https://doi.org/10.1007/3-540-17943-7_135
Show More Cited By

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Media

Figures

Other

Tables

View Issue’s Table of Contents