Abstract
Mapping of non-perfectly nested loops onto a multipipelined architecture is considered. Formulas for the delays between the pipeline starting times are obtained. Graph models of programs (data dependence graph, lattice graph, and computation graph) are used.
Similar content being viewed by others
References
Kalyaev, A.V. and Levin, I.I., Modul’no-narashchivaemye mnogoprotsessornye sistemy so strukturno-protsedurnoi organizatsiei vychislenii (Modular Expandable Multiprocessor Systems with Structural-Procedural Organization of Computations), Moscow: Yanus-K, 2003.
Kalyaev, A.V., Programming of Virtual Parallel Problem-Oriented Supercomputers in the Structure of Universal Massively Parallel Supercomputers, Proc. of the Int. Sci.-Tech. Conf. “Intelligent Multiprocessor Systems”, Taganrog, Sep. 1–5, 1999, pp. 27–39.
Bondalapati, K. and Prasanna, V., Reconfigurable Computing Systems, http://citeseerx.ist.psu.edu/view-doc/summary?doi=10.1.1.11.4918, 2002.
Voevodin, V.V. and Voevodin, Vl.V., Parallel’nye vychislenija (Parallel Computations), St. Petersburg: BHV-Peterburg, 2002.
Voevodin, V.V. and Pakulev, V.V., Determination of Arcs of an Algorithm Graph, Preprint OVM AN SSSR, Moscow, 1989, no. 228.
Feautrier, P., Parametric Integer Programming, RAIRO Recherche Operationnelle, 1988, no. 22, pp. 243–268.
Sergienko, A.M., VHDL dlya proektirovaniya vychislitel’nykh ustroistv (VHDL for Designing Computer Systems), Kiev: Korneichuk, 2003.
Steinberg, R.B., Calculation of Delay Between Pipeline Starting Times for Supercomputers with Structural-Procedural Computing Organization, in Iskusstvennyi intellect, Institute of Artificial Intelligence of Ukrainian National Acad. Sci., Donetsk, DonDIShI, Nauka i Osvita, 2003, no. 4, pp. 105–112.
Steinberg, R.B., Calculation of Delay Between Pipeline Starting Times with account of Data Transfer, Proc. of the Int. Conf. on Parallel Computations and Control Problems (PACO-2006), Moscow: Institute of Control Sciences of the Russian Academy of Sciences, 2006.
Frantsuzov, Yu.A., A Review of Methods for Code Parallelization and Program Pipelining, Programmirovanie, 1992, no. 3, pp. 16–37.
Allen, R. and Kennedy, K., Optimizing Compilers for Modern Architectures, Morgan Kaufmann, 2002.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © R.B. Steinberg, 2010, published in Programmirovanie, 2010, Vol. 36, No. 3.
Rights and permissions
About this article
Cite this article
Steinberg, R.B. Mapping loop nests to multipipelined architecture. Program Comput Soft 36, 177–185 (2010). https://doi.org/10.1134/S0361768810030060
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0361768810030060