• Classified by Research Topic • Sorted by Date • Classified by Publication Type •
ADHA: Automatic Data layout framework for Heterogeneous Architectures.
Deepak Majeti, Kuldeep S. Meel, Raj Barik and Vivek Sarkar.
In Proceedings of Parallel Architecture and Compilation Techniques (PACT), August 2014.
Data layouts play a crucial role in determining the performance of a given application running on a given architecture. Existing parallel programming frameworks for both multicore and heterogeneous systems leave the onus of selecting a data layout to the programmer. Therefore, shifting the burden of data layout selection to optimizing compilers can greatly enhance programmer productivity and application performance. In this work, we introduce ADHA: a two-level hierarchal formulation of the data layout problem for modern heterogeneous architectures. We have created a reference implementation of ADHA in the Heterogeneous Habanero-C (H2C) parallel programming system. ADHA shows significant performance benefits of up to 6.92X compared to manually specified layouts for two benchmark programs running on a CPU+GPU heterogeneous platform.
@inproceedings{MMBS14,
title={ADHA: Automatic Data layout framework for Heterogeneous Architectures},
bib2html_dl_pdf={http://arxiv.org/pdf/1407.4859.pdf},
author={Majeti, Deepak and Meel, Kuldeep S. and Barik, Raj and Sarkar, Vivek},
booktitle=PACT,
year={2014},
month=aug,
bib2html_pubtype={Refereed Conference},
bib2html_rescat={Software Engineering},
abstract={Data layouts play a crucial role in determining the performance of a given application running on a given architecture. Existing parallel programming frameworks for both multicore and heterogeneous systems leave the onus of selecting a data layout to the programmer. Therefore, shifting the burden of data layout selection to optimizing compilers can greatly enhance programmer productivity and application performance. In this work, we introduce ADHA: a two-level hierarchal formulation of the data layout problem for modern heterogeneous architectures. We have created a reference implementation of ADHA in the Heterogeneous Habanero-C (H2C) parallel programming system. ADHA shows significant performance benefits of up to 6.92X compared to manually specified layouts for two benchmark programs running on a CPU+GPU heterogeneous platform.}
}
Generated by bib2html.pl (written by Patrick Riley with layout from Sanjit A. Seshia ) on Thu Aug 22, 2024 18:37:34