Abstract
Parallel applications in a Single Process Multiple Data paradigm assume splitting huge amounts of data to multiple processors working in parallel at small data packets. As the individual data packets are not independent, the processors must interact with each other to exchange results of the calculations with their adjacent partners and take these results into account in their own computations. An example of SPMD is geometric parallelism when a huge map of image data is divided into regular (e.g. rectangular) regions which are assigned to separate processors. The main problem with modeling of such applications in MERPSYS is that a single processor interacts with dynamically pointed partners, but the MERPSYS simulator engine requires static labeling of communication peers. In this chapter we show how we have coped with this problem.
Author (1)
Cite as
Full text
full text is not available in portal
Keywords
Details
- Category:
- Monographic publication
- Type:
- rozdział, artykuł w książce - dziele zbiorowym /podręczniku w języku o zasięgu międzynarodowym
- Title of issue:
- Modeling Large-Scale Computing Systems: Practical Approaches in Merpsys strony 19 - 37
- Language:
- English
- Publication year:
- 2016
- Bibliographic description:
- Kuchta J.: Modeling SPMD Application Execution Time // Modeling Large-Scale Computing Systems: Practical Approaches in Merpsys/ ed. 2016 Gdańsk: Politechnika Gdańska, Wydziałe Elektroniki, Telekomunikacji i Informatyki, Katedra Architektury Systemów Komputerowych, 2016, s.19-37
- Verified by:
- Gdańsk University of Technology
seen 133 times
Recommended for you
Modeling energy consumption of parallel applications
- P. Czarnul,
- J. Kuchta,
- P. Rościszewski
- + 1 authors