Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis (SC); Salt Lake City, UT; Salt Lake City, UT
Год издания: 2012
Идентификатор DOI: 10.1109/SC.Companion.2012.292
Ключевые слова: GPU, CUDA, autotuning, optimization, Accelerator architectures, Autotuner, Block sizes, Magic constants, Optimal values, Digital storage
Аннотация: During GPU-enabled applications optimization a lot of "magic" constants are introduced, such as block size, parallelism granularity, thread weight, and so on. Although these constants affect performance directly, it is impossible to determine their optimal values statistically for the general case because they depend on both the daПоказать полностьюta specificity and the accelerator architecture. This paper describes an approach to these constants tailoring automation that allows to speedup considered benchmarks 20-60 percent at average.
Журнал: 2012 SC COMPANION: HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS (SCC)
Номера страниц: 1508-1509
Место издания: NEW YORK
Издатель: IEEE