Out-of-core cone beam reconstruction using multiple GPUs
Graduate School of Information Science and Technology, Osaka University, 1-5 Yamadaoka, Suita, Osaka 565-0871, Japan
IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2010
@inproceedings{ino2010out,
title={Out-of-core cone beam reconstruction using multiple GPUS},
author={Ino, F. and Okitsu, Y. and Kishi, T. and Ohnishi, S. and Hagihara, K.},
booktitle={Biomedical Imaging: From Nano to Macro, 2010 IEEE International Symposium on},
pages={792–795},
year={2010},
organization={IEEE}
}
This paper presents a graphics processing unit (GPU) based method capable of accelerating cone-beam reconstruction of large volume data, which cannot be entirely stored in video memory. Our method accelerates the Feldkamp, Davis and Kress (FDK) algorithm in a multi-GPU environment. We present how the entire volume can be efficiently decomposed into small portions to reduce the usage of video memory on each graphics card. Experimental results are also presented to understand the reconstruction throughput on an nVIDIA Tesla S1070 server. It takes approximately three minutes to reconstruct a 20483-voxel volume from 720 20482-pixel projections. The effective bandwidth of video memory reaches 137 GB/s per GPU, demonstrating a higher utilization of texture caches.
July 24, 2011 by hgpu
