Wyszukiwarka
Abstrakt
In this work, we evaluate an analytical GPU performance model based on Little's law, that expresses the kernel execution time in terms of latency bound, throughput bound, and achieved occupancy. We then combine it with the results of several research papers, introduce equations for data transfer time estimation, and finally incorporate it into the MERPSYS framework, which is a general-purpose simulator for parallel and distributed systems. The resulting solution enables the user to express a CUDA application in a MERPSYS editor using an extended Java language and then conveniently evaluate its performance for various launch configurations using different hardware units. We also provide a systematic methodology for extracting kernel characteristics, that are used as input parameters of the model. The model was evaluated using kernels representing different traits and for a large variety of launch configurations. We found it to be very accurate for computation bound kernels and realistic workloads, whilst for memory throughput bound kernels and uncommon scenarios the results were still within acceptable limits. We have also proven its portability between two devices of the same hardware architecture but different processing power. Consequently, MERPSYS with the theoretical models embedded in it can be used for evaluation of application performance on various GPUs and used for performance prediction and e.g. purchase decision making.
Cytowania
-
0
CrossRef
-
0
Web of Science
-
1
Scopus
Cytuj jako
Pełna treść
- Wersja publikacji
- Accepted albo Published Version
- DOI:
- Cyfrowy identyfikator dokumentu elektronicznego (otwiera się w nowej karcie) 10.12694/scpe.v19i4.1439
- Licencja
-
otwiera się w nowej karcie
Słowa kluczowe
Informacje szczegółowe
- Kategoria:
- Publikacja w czasopiśmie
- Typ:
- publikacja w in. zagranicznym czasopiśmie naukowym (tylko język obcy)
- Opublikowano w:
-
Scalable Computing: Practice and Experience
nr 19,
wydanie 4,
strony 401 - 422,
ISSN: 1895-1767 - Język:
- angielski
- Rok wydania:
- 2018
- Opis bibliograficzny:
- GAJGER T., Czarnul P.. Modelling and simulation of GPU processing in the MERPSYS environment. Scalable Computing: Practice and Experience, 2018, Vol. 19, iss. 4, s.401-422
- DOI:
- Cyfrowy identyfikator dokumentu elektronicznego (otwiera się w nowej karcie) 10.12694/scpe.v19i4.1439
- Weryfikacja:
- Politechnika Gdańska
wyświetlono 146 razy
Publikacje, które mogą cię zainteresować
Modeling energy consumption of parallel applications
- P. Czarnul,
- J. Kuchta,
- P. Rościszewski
- + 1 autorów