Abstract
Starting from categories of the known computer-networks-area like message-handling, bus, protocol etc. may lead to a dead-end with respect to multiprocessor-design. Those terms from the world of networks may persuade computer architects to choose solutions which are not adequate to the requirements of efficient and fault-tolerant operation.
Topological investigations including technological considerations will result in structures which differ contrarily from the traditional monoprocessor as well as from computer networks. The macro-dataflow-concept [11] for instance can ensure in the framework of the usual storage-access a high rate of efficiency. Also, system programming and application programming will not deviate in this case too much from the traditional patterns. Beyond it the approaches for fault-tolerant operation become then simple and effective.
Experiences with multiprocessors, which are accordingly designed are discussed.
Preview
Unable to display preview. Download preview PDF.
References
Enslow, Philip H. edit.: Multiprocessors & Parallel Processing; John Wiley & Sons, New York, London etc. 1974
Händler, W.: Dynamic Computer Structures for Manifold Utilization; Parallel Computing 2 (1985), 15–32
Händler, W.: Bericht über eine Dienstreise nach Japan in der Zeit vom 31. Mai bis 15. Juni 1986; Interner Bericht, hektographiert Erlangen 1986
Händler, W.; F. Hofmann and H. J. Schneider: A general purpose array with a broad spectrum of applications, Proc. Workshop on Computer Architecture IFB (Springer, Berlin, 1975) 311–335 (und zahlreiche andere Veröffentlichungen zum Projekt EGPA)
z. B. auch: Händler, W.; U. Herzog, F. Hofmann and H. J. Schneider: Multiprozessoren für breite Anwendungsbereiche — Erlangen general purpose array, Conf. Proc. 8. GI-NTG Fachtagung ‘Architektur und Betrieb von Rechensystemen' (Springer, Berlin, 1984) 195–208
Händler, W.; H. Rohrer: Gedanken zu einem Rechner-Baukastensystem; Elektronische Rechenanlagen, 22 (1980), 3–13 (sowie weitere Veröffentlichungen zum Projekt DIRMU)
z. B. auch: Händler, W.; E. Maehle and K. Wirl: The DIRMU Testbed for High-Performance Multiprocessor Configurations; Proc. of the First International Conference on Supercomputing Systems (Dec. 16–20, 1985, St. Petersburg, FL.) 468–475
Hoshino, T.: An Invitation to the World of PAX (using contemporary LSI and simple architecture, we can construct a multiple-purpose machine PAX with a 1 Gflops speed. Our Goal for the 1990's: a 1 Tflops PAX); Computer, May 1986, 68–79
oder: Momoi, S.; S. Shinada, M. Kobayashi and T. Ishikawa: Hierarchical Array Processor System (HAP); Proceeding CONPAR '86 Conference on Algorithms and Hardware for Parallel Processing Aachen, September 1986, Händler et al. (edit.), vol. 237, Lecture Notes in Computer Science, Springer-Verlag Berlin, Heidelberg, New York etc. 1986
Albert B.; A. Bode, W. Händler: A case study in vertical migration: The implementation of a dedicated associative instruction set; in: Microprocessing and Mikroprogramming 8 / 1981, 257–262 (Schriftum siehe auch [2]).
Henning, W.; J. Volkert: Programming EGPA-Systems; Proc. 5th Intern. Conf. on Distributed Computing Systems, Denver 1985, 552–558
MCC: Multi-Chip-Carrier — compare sources of IBM. A Ceramic-carrier allows to combine a multitude of chips on a minimal space. The chips are mutually connected by a direct bonding.
Händler, W.: Multiprozessoren: Effizienz und Fehlertoleranz; NTG Fachberichte 92, VDE-Verlag GmbH 1986
Händler, W.: The concept of Macro-Pipelining with high availability. In: Eletronische Rechenanlagen 6, 1973, 269–274
Galil, Z.; W. Paul: Effizienz paralleler Rechner. In: IFB 33, 10. GI-Jahrestagung, Springer-Verlag Berlin, Heidelberg, New York 1980
Händler, W.; G. Fritsch, J. Volkert: Applications implemented on the Erlangen General Purpose Array. (main lecture) 12–31. In: Parcella '84, Proceedings of the II. Intern. Workshop on Parallel Processing by Cellular Automate and Arrays (Berlin, Sept. 25–27, 1984)
Smith, B.J.: A pipelined shared resource MIMD computer. In: Proceedings of the 1978 Intern. Conference on Parallel Processing, 6–8, IEEE, New York
Maehle, E.; K. Moritzen and K. Wirl: A Graph Model and its Application to a Fault-Tolerant Multiprocessor System. Proc. 16th Intern. Symp. on Fault-Tolerant Computing ‘FTCS-16', Wien 1986, 292–297
or: Maehle, E.; H. Joseph: Selbstdiagnose in fehlertoleranten DIRMU-Multi-Mikroprozessorkonfigurationen. In: Fehlertolerante Rechensysteme, Informatik-Fachberichte Vol. 55 1982
Patterson, D. A.; C. H. Sequin: A VLSI RISC. IEEE Computer 15 (1982) No. 9 (Sept.) 8–21
or e.g.: Horster, P.; D. Manstetten, H. Pelzer: RISC, Reduced Instruction Set Computer. Konzepte und Realisierungen, Hüthig-Verlag, Heidelberg 1987
Fritsch, G.: Memory-Coupled Arrays for a Broad Spectrum of Applications. Proc. WOPPLOT 83, Parallel Processing: Logic, Organization and Technology, Springer-Verlag Berlin, Heidelberg, New York, Tokyo 1984, 158–177
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1989 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Händler, W. (1989). Multiprocessor arrays: Topology, efficiency and fault-tolerance. In: Wolf, G., Legendi, T., Schendel, U. (eds) Parcella '88. Parcella 1988. Lecture Notes in Computer Science, vol 342. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-50647-0_100
Download citation
DOI: https://doi.org/10.1007/3-540-50647-0_100
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-50647-8
Online ISBN: 978-3-540-46062-6
eBook Packages: Springer Book Archive