Combining data reuse exploitation with data-level parallelization for FPGA targeted hardware compilation: A geometric programming framework
Uložené v:
| Názov: | Combining data reuse exploitation with data-level parallelization for FPGA targeted hardware compilation: A geometric programming framework |
|---|---|
| Autori: | Qiang Liu, George A. Constantinides, Konstantinos Masselos, Peter Y. K. Cheung |
| Prispievatelia: | The Pennsylvania State University CiteSeerX Archives |
| Zdroj: | http://cas.ee.ic.ac.uk/people/gac1/pubs/QiangFPL08.pdf. |
| Rok vydania: | 2008 |
| Zbierka: | CiteSeerX |
| Popis: | A geometric programming framework is proposed in this paper to automate exploration of the design space consisting of data reuse (buffering) exploitation and loop-level parallelization, in the context of FPGA-targeted hardware compilation. We expose the dependence between data reuse and data-level parallelization and explore both problems under the on-chip memory constraint for performance-optimal designs within a single optimization step. Results from applying this framework to several real benchmarks demonstrate that given different constraints on on-chip memory utilization, the corresponding performance-optimal designs are automatically determined by the framework, and performance improvements up to 4.7 times have been achieved compared with the method that first explores data reuse and then performs parallelization. 1. |
| Druh dokumentu: | text |
| Popis súboru: | application/pdf |
| Jazyk: | English |
| Relation: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.385.9818; http://cas.ee.ic.ac.uk/people/gac1/pubs/QiangFPL08.pdf |
| Dostupnosť: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.385.9818 http://cas.ee.ic.ac.uk/people/gac1/pubs/QiangFPL08.pdf |
| Rights: | Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
| Prístupové číslo: | edsbas.26329790 |
| Databáza: | BASE |
Nájsť tento článok vo Web of Science | Abstrakt: | A geometric programming framework is proposed in this paper to automate exploration of the design space consisting of data reuse (buffering) exploitation and loop-level parallelization, in the context of FPGA-targeted hardware compilation. We expose the dependence between data reuse and data-level parallelization and explore both problems under the on-chip memory constraint for performance-optimal designs within a single optimization step. Results from applying this framework to several real benchmarks demonstrate that given different constraints on on-chip memory utilization, the corresponding performance-optimal designs are automatically determined by the framework, and performance improvements up to 4.7 times have been achieved compared with the method that first explores data reuse and then performs parallelization. 1. |
|---|