Authors

V.A. Bakhtin, N.A. Kataev, A.S. Kolganov, N.N. Korolev, V.A. Krukov, M.Yu. Kuznecov, N.V. Podderugina, M.N. Pritula, O.A. Savitskaya, A.A. Smirnov, O.F. Zhukova

Annotation

The experience of SAPFOR system usage  has shown that when parallelizing large programs and program systems for a cluster, it is necessary to  parallelize them incrementally, starting from the most time-consuming fragments and incrementally adding new fragments, until we reach the desired level of  the parallel program performance. The principles of incremental parallelization of the program systems are considered in the article.

The work was funded by RFBR (projects No. 16-07-01067, 16-07-01014, 17-01-00820).

Key words

Automation of parallelization, heterogeneous computational cluster, incremental parallelization.

Presented on

XIX all_Russian scientific conference Scientific service on the Internet 2017