Evaluation of Autoparallelization Toolkits for Commodity GPUs

Evaluation of Autoparallelization Toolkits for Commodity GPUs
复制标题

商用 GPU 自动并行化工具包的评估

DOI:
10.1007/978-3-642-55224-3_42
复制
发表时间:
2013
影响因子:
3.3
通讯作者:
J. Roerdink
J. Roerdink
中科院分区:
物理与天体物理2区
文献类型:
--
作者:
David Williams;V. Codreanu;Po;Baoquan Liu;Feng Dong;Burhan Yasar;Babak Mahdian;A. Chiarini;Xia Zhao;J. Roerdink

文献摘要

被引文献

相似文献

在本文中,我们评估了OpenACC和Mint工具包的性能,以针对标准PolyBench测试套件的C和CUDA实现。我们的分析表明,在许多情况下,性能是相似的,但是一套代码构造会阻碍薄荷生成最佳代码的能力。然后,我们提出了一些小改进,这些改进将我们集成到自己的GPSME工具包中(源自MINT),并表明我们的工具包现在在大多数测试中都超过了OpenACC。
In this paper we evaluate the performance of the OpenACC and Mint toolkits against C and CUDA implementations of the standard PolyBench test suite. Our analysis reveals that performance is similar in many cases, but that a certain set of code constructs impede the ability of Mint to generate optimal code. We then present some small improvements which we integrate into our own GPSME toolkit (which is derived from Mint) and show that our toolkit now out-performs OpenACC in the majority of tests.