TARGETING HETEROGENEOUS ARCHITECTURES VIA MACRO DATA FLOW
Abstract
We propose a data flow based run time system as an efficient tool for supporting execution of parallel code on heterogeneous architectures hosting both multicore CPUs and GPUs. We discuss how the proposed run time system may be the target of both structured parallel applications developed using algorithmic skeletons/parallel design patterns and also more "domain specific" programming models. Experimental results demonstrating the feasibility of the approach are presented.
This work has been supported by European Union Framework 7 grant IST-2011-288570 "ParaPhrase: Parallel Patterns for Adaptive Heterogeneous Multicore Systems".
References
-
Timothy Mattson , Beverly Sanders and Berna Massingill , Patterns for parallel programming , 1st edn. ( Addison-Wesley Professional , 2004 ) . Google Scholar - Parallel Computing 30(3), 389 (2004). Crossref, ISI, Google Scholar
- Softw., Pract. Exper. 40(12), 1135 (2010). Crossref, ISI, Google Scholar
- ACM Comput. Surv. 36, 1 (2004). Crossref, ISI, Google Scholar
Jack B. Dennis and David P. Misunas , A preliminary architecture for a basic data-flow processor, Proceedings of the 2nd annual symposium on Computer architecture, ISCA '75 (ACM, New York, NY, USA, 1975) pp. 126–132. Google Scholar- Commun. ACM 28(1), 34 (1985). Crossref, ISI, Google Scholar
- Parallel Processing Letters 11(1), 41 (2001). Link, Google Scholar
- SIGARCH Comput. Archit. News 36, 29 (2009). Crossref, Google Scholar
Samer Arandi and Paraskevas Evripidou , Programming multi-core architectures using Data-Flow techniques, ICSAMOS'10 (2010) pp. 152–161. Google ScholarPritish Jetley and Laxmikant V. Kal , Static macro data flow: Compiling global control into local control, IPDPS Workshops'10 (2010) pp. 1–8. Google Scholar- Concurrency Practice and Experience 7(3), 225 (1995). Crossref, ISI, Google Scholar
Susanna Pelagatti , Task and data parallelism in P3L (Springer-Verlag, London, UK, 2003) pp. 155–186. Crossref, Google Scholar-
Gerhard R. Joubert (eds.) , Parallel Computing: Current & Future Issues of High-End Computing, Proceedings of the International Conference ParCo 2005 33 ,John von Neumann Institute for Computing Series ( Central Institute for Applied Mathematics , Jülich, Germany , 2005 ) . Google Scholar - Concurrency and Computation: Practice and Experience 22, 15 (2010). Crossref, ISI, Google Scholar
M. Aldinucci , Parallel patterns + Macro Data Flow for multi-core programming, Proceedings of the 20th International EuroMicro Conference on Parallel, Distributed and Network-based Processing (Conference Publishing Services IEEE, 2012) pp. 27–36. Google ScholarJohan Enmyren and Christoph W. Kessler , Skepu: a multi-backend skeleton programming library for multi-gpu systems, Proceedings of the fourth international workshop on High-level parallel programming and applications, HLPP '10 (ACM, New York, NY, USA, 2010) pp. 5–14. Google Scholar-
S. Ernsting and H. Kuchen , Data Parallel Skeletons for GPU Clusters and Multi-GPU Systems , Proceedings of PARCO 2012 ( Gent , 2012 ) . Google Scholar Michel Steuwer , Philipp Kegel and Sergei Gorlatch , Skelcl - a portable skeleton library for high-level gpu programming, Proceedings of the 2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and PhD Forum, IPDPSW '11 (IEEE Computer Society, Washington, DC, USA, 2011) pp. 1176–1182. Google Scholar-
U. Dagstgeer , C. Kessler and S. Thibault , Flexible runtime support for efficient skeleton programming on hybrid systems , Proceedings of PARCO 2012 ( Gent , 2012 ) . Google Scholar - IEEE Trans. Parallel Distrib. Syst. 20, 404 (2009). Crossref, ISI, Google Scholar
- R. Badia. StarSs support for programming heterogeneous platforms. In Proceedings of Heteropar 2012 workshop, to appear in EuroPar 2012 Parallel Computing Workshop Proceedings, 2012 . Google Scholar
- IJHPCA 23(3), 284 (2009). ISI, Google Scholar
Marco Aldinucci , Marco Danelutto and Peter Kilpatrick , Towards hierarchical management of autonomic components: a case study, Proc. of Intl. Euromicro PDP 2009: Parallel Distributed and network-based Processing, eds.Didier El Baz , Tom Gross and Francois Spies (IEEE, Weimar, Germany, 2009) pp. 3–10. Google Scholar


