Abstract
An approach of dynamical parallelization of computation and communication based on algebraic treatment of programs and advanced transition-system operational semantics is described parallel programs with distributed/shared memory. The approach combines both data flow analysis and buffering techniques to improve communication and synchronization features for these programs. This is accomplished in two ways: 1) construction a number of semantic models of data exchanges of increasing power for asynchronous communications and 2) exploiting user-based specification of shared memory access order by means of orthogonal synchronization facilities that are more expressive and efficient than barrier-like ones. Two new programming abstractions aimed to reduce communication and synchronization overhead are elaborated from the theory and shown to have immediate practical influence on parallel programming.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
R.H. Campbell, N.A. Habermann, The specification of process synchronization by path expressions, in: Lect. Notes Comput. Sci., 1974, vol.16, p. 89–102.
A. Dinning, E. Schonberg, An empirical comparison of monitoring algorithms for access anomaly detection, SIGPLAN Notices, 25, March 1990, pp. 1–10.
A.E. Doroshenko, Enhancing Asynchronism of Data Exchanges in Parallel Programs, in Parallel Computing Technologies (PaCT-93), Proc. Int. Conf., ed. V.E. Malyshkin, NT-Center, Moscow, Russia, 1993, pp. 291–300.
A.E Doroshenko, On asynchronous avoiding deadlocks in parallel programs, Parallel Processing Letters, 1992, 2, No. 2–3, pp. 291–297.
A.E. Doroshenko, Advancing synchronization and communication techniques for distributed/shared memory parallel programs, in: PARCELLA '94: Proc. VI Int. Workshop on Parallel Processing by Cellular Arrays and Automata, Academie Verlag, Berlin, 1994, pp. 131–139.
A. E. Doroshenko, Programming Abstracts for Synchronization and Communication in Parallel Programs, in: V. Malyshkin, ed., Parallel Computing Technologies, Proc. Third Int. Conf. PaCT'95, Lect. Notes Comput. Sci., vol. 964 (Springer, 1995) 157–162.
A. E. Doroshenko, A. B. Godlevsky, Parallelizing Programs with Algebraic Programming Tools, in: S. Haridi, K. Ali, P. Magnusson, eds., EURO-PAR'95: Parallel Processing. Proc. Int. Conf., Lect. Notes Comput. Sci., vol. 966 (Springer, 1995) 687–690.
D. Gelernter, N. Carriero, Coordination Languages and Their Significance, Commun. ACM 35, No. 2 (1992) 97–107.
C.A.R. Hoare, Communicating sequential processes, Commun. ACM 21, No. 8 (1978) 666–677.
R.M. Keller, A fundamental theorem of asynchronous parallel computations, Lect. Notes in Comput. Sci. 24 (1975) 102–112.
MPI: A Message-Passing Interface Standard, Int. J. of Supercomputer Applications and High Performance Computing, 8, No. 3/4, Fall/Winter 1994, pp. 159–416.
A.N. Refenes, Message passing via single buffered channels: an efficient and flexible communication control mechanism, Microprocessing and Microprogramming, 30, (1990) 645–654.
N.B. Serbedzija, Asynchronous communication in OCCAM, SIGPLAN Notices 24 (1988) 51–62.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Doroshenko, A.E. (1997). Modeling synchronization and communication abstractions for dynamical parallelization. In: Hertzberger, B., Sloot, P. (eds) High-Performance Computing and Networking. HPCN-Europe 1997. Lecture Notes in Computer Science, vol 1225. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0031646
Download citation
DOI: https://doi.org/10.1007/BFb0031646
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-62898-9
Online ISBN: 978-3-540-69041-2
eBook Packages: Springer Book Archive