11331

Comparing the Performance of Different x86 SIMD Instruction Sets for a Medical Imaging Application on Modern Multi- and Manycore Chips

Johannes Hofmann, Jan Treibig, Georg Hager, Gerhard Wellein
Chair for Computer Architecture, University Erlangen-Nuremberg
arXiv:1401.7494 [cs.DC], (29 Jan 2014)

@article{2014arXiv1401.7494H,

   author={Hofmann, Johannes and Treibig, Jan and Hager, Georg and Wellein, Gerhard},

   title={Comparing the Performance of Different x86 SIMD Instruction Sets for a Medical Imaging Application on Modern Multi- and Manycore Chips},

   journal={ArXiv e-prints},

   archivePrefix={"arXiv"},

   eprint={1401.7494},

   primaryClass={"cs.DC"},

   keywords={Distributed, Parallel, and Cluster Computing,Performance},

   year={2014},

   month={jan}

}

Download Download (PDF)   View View   Source Source   

2436

views

Single Instruction, Multiple Data (SIMD) vectorization is a major driver of performance in current architectures, and is mandatory for achieving good performance with codes that are limited by instruction throughput. We investigate the efficiency of different SIMD-vectorized implementations of the RabbitCT benchmark. RabbitCT performs 3D image reconstruction by back projection, a vital operation in computed tomography applications. The underlying algorithm is a challenge for vectorization because it consists, apart from a streaming part, also of a bilinear interpolation requiring scattered access to image data. We analyze the performance of SSE (128 bit), AVX (256 bit), AVX2 (256 bit), and IMCI (512 bit) implementations on recent Intel x86 systems. A special emphasis is put on the vector gather implementation on Intel Haswell and Knights Corner microarchitectures. Finally we discuss why GPU implementations perform much better for this specific algorithm.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: