A Comprehensive Performance Comparison of CUDA and OpenCL

By Jianbin Fang, Ana Lucia Varbanescu, Henk J. Sips. International Conference on Parallel Processing, 2011

322 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/d86f5b4ad32d2daa3d8223e10994125b1298de5b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

1. Anders Eklund
2. Graham W. Taylor
3. Luke Smith
4. Daniel Forsberg
5. Paul Dufort
6. Stephen LaConte
7. Qiuhua Liang

In Hardware and Architecture

1. Graham W. Taylor
2. Luke Smith
3. Qiuhua Liang
4. Shimeng Yu
5. Wei Yu
6. William Grant Hatcher

Citations in the corpus

• Medical image processing on the GPU - Past, present and future (2013, methodology): “, 2011), but this is diminishing as OpenCL compilers improve and now typically stands at no more than 20%-30% (Fang et al., 2011).”
• Deep Learning on FPGAs: Past, Present, and Future (2016, background): “While a detailed comparison of OpenCL and CUDA is outside the scope of this paper, performance of both frameworks has been shown to be very similar for given applications [22].”
• Towards a generalised GPU/CPU shallow-flow modelling tool (2013, methodology): “The software developed herein uses OpenCL to provide the greatest compatibility between different device and hardware configurations, whilst the performance levels achievable are comparable to popular alternatives such as CUDA Fang et al. (2011).”