Write a Blog >>
SPLASH 2017
Sun 22 - Fri 27 October 2017 Vancouver, Canada
Wed 25 Oct 2017 10:52 - 11:15 at Regency C - Performance Chair(s): Kathryn S McKinley

Heterogeneous architectures characterize today hardware ranging from super-computers to smartphones. However, in spite of this importance, programming such systems is still challenging. In particular, it is challenging to map computations to the different processors of a heterogeneous device. In this paper, we provide a static analysis that mitigates this problem. Our contributions are two-fold: first, we provide a semi-context-sensitive algorithm, which analyzes the program's call graph to determine the best processor for each calling context. This algorithm is parameterized by a cost model, which takes into consideration processor's characteristics and data transfer time. Second, we show how to use simulated annealing to calibrate this cost model for a given heterogeneous architecture. We have used our ideas to build Etino, a tool that annotates C programs with OpenACC or OpenMP 4.0 directives. Etino generates code for a CPU-GPU architecture without user intervention. Experiments on classic benchmarks reveal speedups of up to 75x. Moreover, our calibration process lets \etino{} avoid slowdowns of up to 720x which trivial parallelization approaches would yield.

Wed 25 Oct

splash-2017-OOPSLA
10:30 - 12:00: OOPSLA - Performance at Regency C
Chair(s): Kathryn S McKinleyGoogle
splash-2017-OOPSLA150892020000010:30 - 10:52
Talk
Lun LiuUniversity of California at Los Angeles, USA, Todd MillsteinUniversity of California, Los Angeles, Madan MusuvathiMicrosoft Research
DOI
splash-2017-OOPSLA150892155000010:52 - 11:15
Talk
Gabriel PoesiaFederal University of Minas Gerais, Brazil, Breno CamposFederal University of Minas Gerais, Brazil, Fabrício FerracioliLG Electronics, Brazil, Fernando Magno Quintão PereiraUFMG
DOI
splash-2017-OOPSLA150892290000011:15 - 11:37
Talk
Sarah ChasinsUniversity of California, Berkeley, Rastislav BodikUniversity of Washington
DOI
splash-2017-OOPSLA150892425000011:37 - 12:00
Talk
Edd BarrettKing's College London, Carl Friedrich Bolz-Tereick, Rebecca KillickDepartment of Mathematics and Statistics, University of Lancaster, Sarah MountKing's College London, Laurence TrattKing's College London
DOI