Abstract
In this paper we present MaDCoWS, a software implementation of a Distributed Shared Memory (DSM) runtime system, specifically designed for massively parallel 2-D grid multiprocessors. The system takes advantage of the network topology in order to minimise the paths of the message sequences realising the shared operations. As a result its performance is increased and the system becomes scalable even to very large processor numbers. We present the basic ideas for 2-D optimisations, the implementation structure and results from synthetic and application benchmarks executed on a 1024 processor Parsytec GCel.
This work was partly supported by the Greek General Secretariat of Research and Technology under the YPER-94 program.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
Parsytec Computer GmbH, ‘PARIX 1.2: User Manual', 1993.
Cordsen, J. et al., W. ‘On the Coexistence of Shared-Memory and Message Passing in the Programming of Parallel Applications'. In Proc. of the HPCNE'97, pp. 718–727, Apr 1997.
Freeh, V. W. and Andrews, G. R. ‘Dynamically Controlling False Sharing in Distributed Shared Memory'. In Proc. of the Fifth IEEE Int'l Symp. on High Performance Distributed Computing (HPDC-5), pages 403–411, August 1996.
Lamport, L. ‘How to Make a Multiprocessor Computer That Correctly Executes Multiprocess programs'. IEEE Transactions on Computers, C-28(9):690–691, Sept. 1979.
Gharachorloo, K., Gupta, A., and Hennessy, J. L. Revision to ‘Memory Consistency and Event Ordering in Scalable Shared-Memory Multiprocessors'. Technical Report CSL-TR-93-568, Computer Systems Laboratory, Stanford University, April 1993.
Dimitrelos, D and Halatsis, C. ‘Improving the Performance of Distributed Shared Memory Environments on Grid Multiprocessors', Submitted to EUROPAR'99.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1999 Springer-Verlag
About this paper
Cite this paper
Dimitrelos, D., Halatsis, C. (1999). MaDCoWS: A scalable distributed shared memory environment for massively parallel multiprocessors. In: Sloot, P., Bubak, M., Hoekstra, A., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1999. Lecture Notes in Computer Science, vol 1593. Springer, Berlin, Heidelberg . https://doi.org/10.1007/BFb0100639
Download citation
DOI: https://doi.org/10.1007/BFb0100639
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65821-4
Online ISBN: 978-3-540-48933-7
eBook Packages: Springer Book Archive