Cache-aware GPU memory scheduling scheme for CT back-projection
Cache-aware GPU memory scheduling scheme for CT back-projection
复制标题
CT反投影的缓存感知GPU内存调度方案
DOI：
10.1109/nssmic.2010.5874183
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
K. Mueller
中科院分区：
文献类型：
--
作者：
Ziyi Zheng;K. Mueller
图形处理单元（GPU）非常适合计算密集型任务，是执行计算机断层扫描（CT）重建的最快解决方案之一。正如以前的研究表明，GPU实现的瓶颈不是计算能力，而是内存带宽。我们提出了一个缓存感知的内存调度方案的反投影，它可以确保更好的GPU处理器和GPU内存之间的负载平衡。所提出的重组方法可以直接应用于现有的GPU加速CT重建管道。实验结果表明，我们的优化可以达到1.18-1.48的加速比。我们的缓存优化方法对于具有高分辨率投影的低分辨率卷特别有效。
Graphic process units (GPUs) are well suited to computing-intensive tasks and are among the fastest solutions to perform Computed Tomography (CT) reconstruction. As previous research shows, the bottleneck of GPU-implementation is not the computational power, but the memory bandwidth. We propose a cache-aware memory-scheduling scheme for the back-projection, which can ensure a better load-balancing between GPU processors and the GPU memory. The proposed reshuffling method can be directly applied on existing GPU-accelerated CT reconstruction pipelines. The experimental results show that our optimization can achieve speedup ranging from 1.18–1.48. Our cache-optimization method is particular effective for low-resolution volumes with high resolution projections.
影响因子：
1.4
作者：
Okitsu, Yusuke;Ino, Fumihiko;Hagihara, Kenichi
通讯作者：
Hagihara, Kenichi