8108

Is OpenCL a suitable platform for algorithm development in health care systems?

Mattias Larsson
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology
Uppsala University, 2012
@phdthesis{larsson2012opencl,

   title={Is OpenCL a suitable platform for algorithm development in health care systems?},

   author={Larsson, M.},

   year={2012},

   school={Uppsala University}

}

Download Download (PDF)   View View   Source Source   

1007

views

This thesis reviews if OpenCL is a suitable and cost effective platform for algorithm development in health care systems. Aspects such as maintainability, performance, portability and integration with high-level languages (in this case Python) are analyzed. The review is done by implementing one part of a dose calculation algorithm that is complex enough to provide a realistic case. The vision is that OpenCL can replace multiple platforms for both multi core CPU and GPU computing and removing the need of implementing an optimized version of an algorithm for every platform. To achieve performance-portability, automatic optimization is done using parameter tuning. Both its effects on performance and code structure are analyzed. The conclusion is that OpenCL coupled with auto tuning is not a suitable platform due to problems with code structure, language limitations, programming-portability, tool support and the effort and difficulty in implementing auto tuning.
VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
  • He is a Python-programmer and finds choice for C in the kernels limited. I think he came to good conclusions as OpenCL should advance in these points. Bus his conclusion feels like saying CUDA is not fit for health care, as it does not run on Radeons. These things have little causality.

    The algorithms needed in health care can be sped up with GPUs (being either CUDA or OpenCL). Instead of above conclusion I would like to see a conclusion on how much these algorithms can be sped up and how useful this speed-up would be.

* * *

* * *

Follow us on Twitter

HGPU group

1946 peoples are following HGPU @twitter

Like us on Facebook

HGPU group

443 people like HGPU on Facebook

HGPU group © 2010-2016 hgpu.org

All rights belong to the respective authors

Contact us: