节点文献
基于GPU实现叠前时间偏移走时计算的并行算法
Parallel Algorithm Based on the Travel Time Computing of Pre-stack Time Migration Using GPU
【摘要】 走时计算是叠前时间偏移计算中最耗时的部分,通过分析传统的串行走时算法,发现静态8点插值算法非常适合在GPU上运行。首先利用CUDA技术对静态8点插值算法进行并行化改造,设计静态8点并行插值算法,然后测试其正确性,统计其相对误差情况。实验表明此算法比工业生产上的动态插值算法更准确,最后我们利用体偏作性能测试。试验结果表明,运行在GPU上的静态8点并行插值算法内核性能是运行在CPU上的动态插值算法内核的22.76倍。这说明,静态8点并行插值算法适合进行走时计算,并且可以应用于工业生产上。
【Abstract】 The computation of traveling time of Pre-Stack Time Migration(PSTM) is the most time-consuming part of whole procedure.In this paper,we present a novel parallel algorithm to low the time of traveling-time computation.By using static interpolation with step of 8 points instead of dynamic step length,the new designed algorithm performs more accuracy and performance.Experimental results show that the speed-up number reaches 22.76 times.The statistical relative error demonstrates the better accuracy comparing with the commercial codebases.
【Key words】 PSTM; seismic exploration; GPU; CUDA; the travel time computing; parallel computing;
- 【文献出处】 计算机系统应用 ,Computer Systems & Applications , 编辑部邮箱 ,2011年08期
- 【分类号】TP338.6
- 【被引频次】7
- 【下载频次】106