PARTANS:An autotuning framework for stencil computation on multi-GPU systems
Autor: | Thibaut Lutz, Murray Cole, Christian Fensch |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2013 |
Předmět: |
020203 distributed computing
Computer science Stencil code 02 engineering and technology Parallel computing Stencil Operator (computer programming) Hardware and Architecture Homogeneous 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Multi gpu General-purpose computing on graphics processing units Focus (optics) Software Information Systems PCI Express ComputingMethodologies_COMPUTERGRAPHICS |
Zdroj: | Lutz, T, Fensch, C & Cole, M 2013, ' PARTANS : An autotuning framework for stencil computation on multi-GPU systems ', ACM Transactions on Architecture and Code Optimization, vol. 9, no. 4, 59 . https://doi.org/10.1145/2400682.2400718 ACM Transactions on Architecture and Code Optimization |
Popis: | GPGPUs are a powerful and energy-efficient solution for many problems. For higher performance or larger problems, it is necessary to distribute the problem across multiple GPUs, increasing the already high programming complexity. In this article, we focus on abstracting the complexity of multi-GPU programming for stencil computation. We show that the best strategy depends not only on the stencil operator, problem size, and GPU, but also on the PCI express layout. This adds nonuniform characteristics to a seemingly homogeneous setup, causing up to 23% performance loss. We address this issue with an autotuner that optimizes the distribution across multiple GPUs. |
Databáze: | OpenAIRE |
Externí odkaz: |