Automated Development of Applications for Graphical Processing Units Using Rewriting Rules |
Anatoliy Doroshenko, Konstantin Zhereb |
Abstract
Recently there was an active development of parallel programming methods concerning implementation of general-purpose algorithms on graphical processing units (GPUs). Using this specialized hardware allows increasing performance significantly, but requires low-level programming and understanding details of underlying hardware and software platform. Therefore there is a need for automating development process. This paper presents a technique for automating GPU application development, based on rewriting rules approach. An example is given demonstrating the possibilities of our approach when migrating from sequential C# program to its parallel analog running on GPU, as well as optimization of parallel applications. Using our approach we obtained performance speedup of 25X, while preserving the benefits of Microsoft .NET platform. |
Bio Anatoliy Yu. Doroshenko – Professor of the Department of Automation and Control of National University of Technology “Kyiv Polytechnic Institute”, Head of Department of Theory of Computing in the Institute of Software Systems of National Academy of Sciences of Ukraine (part time) and visiting Professor at the Department of Programming Theory and Technology of Kyiv National University by Taras Shevchenko (KNU).
|
Bio Konstantin Zhereb obtained his Master degree in Applied Mathematics and Physics from Moscow Institute of Physics and Technology (MIPT). He is currently a PhD student at Kyiv Branch of MIPT (Scientific Educational Center of Physics and Technology of National Academy of Sciences of Ukraine). He is also R&D lead at INTSPEI. Konstantin is Microsoft Certified Professional Developer (MCPD: Windows Developer). |
« Previous Regular talk | All Regular talks | Next Regular talk » |
Keynotes | Invited talks | Workshops | Types of presentations |
All talks sorted by author name | All talks sorted by company name |