摘要
为了适应大规模地震勘探海量数据快速处理的需求,本文深入分析了地震勘探数据处理中广泛应用的Kirchhoff叠前时间偏移方法,提出了一种利用Hadoop技术实现叠前时间偏移的并行算法;采用数据的两次分块和内存循环利用等技术,提高了Kirchhoff叠前时间偏移海量数据处理效率。实际数据的测试表明,基于Hadoop的叠前时间偏移算法的计算效率比MPI版本提高了6.7%,同时在大规模生产应用条件下(大规模工区和大规模集群系统)具有更高的稳定性、容错性和适应性。
In order to meet the requirement of the massive data rapid processing during large-scale seismic exploration,we analyze in this article Kirchhoff prestack time migration method which is widely used in seismic data processing,and put forward aparallel algorithm based on Hadoop to achieve prestack time migration.Adopting dataset double partition and memory recycling,Kirchhoff prestack time migration efficiency on the massive data processing could be improved.Tests on real data show that the efficiency of the proposed Kirchhoff prestack time migration algorithm is 6.7% higher than MPI version.And furthermore this algorithm has higher stability,higher fault-tolerance,and higher adaptability under the conditions of the mass production application(lager-scale work area and large-scale cluster system)
出处
《石油地球物理勘探》
EI
CSCD
北大核心
2015年第6期1213-1218,1036,共6页
Oil Geophysical Prospecting
