首页 | 本学科首页   官方微博 | 高级检索  
     

基于网格技术的地震资料并行处理平台建设与应用
引用本文:钟 敏,陈朝根,葛宇飞,亓雪冬,梁 鸿,仝兆岐. 基于网格技术的地震资料并行处理平台建设与应用[J]. 中国石油大学学报(自然科学版), 2014, 0(2): 180-186
作者姓名:钟 敏  陈朝根  葛宇飞  亓雪冬  梁 鸿  仝兆岐
作者单位:中国石油大学计算机与通信工程学院;中国石油大学地球科学与技术学院
基金项目:中国教育科研网格Chinagrid项目;中央高校基本科研业务费专项资金项目(R1107009A)
摘    要:针对地球物理勘探领域海量数据处理需求和行业高性能资源整合需求,结合地震资料数据并行处理特征,利用网格技术建设应用网格分布并行处理平台,详细介绍平台的体系结构和关键技术,成功部署包含两个虚拟社区的应用网格平台。进行Marmousi模型地震波场正演模拟网格并行处理、基于模糊聚类作业划分策略的叠前深度偏移并行处理、积分法叠前时间偏移并行处理。验证了平台的稳定性和并行作业划分策略的有效性,平台效率与传统并行处理平台相比效率相当。结果表明,利用该平台可以进一步整合更多高性能资源,扩大并行处理规模,提高资源的利用率,缩短数据处理周期。

关 键 词:网格  分布并行  元调度  地震资料处理
收稿时间:2013-11-11

Gird platform for seismic data parallel processing and its application
ZHONG Min,CHEN Chao-gen,GE Yu-fei,QI Xue-dong,LIANG Hong and TONG Zhao-qi. Gird platform for seismic data parallel processing and its application[J]. Journal of China University of Petroleum (Edition of Natural Sciences), 2014, 0(2): 180-186
Authors:ZHONG Min  CHEN Chao-gen  GE Yu-fei  QI Xue-dong  LIANG Hong  TONG Zhao-qi
Affiliation:ZHONG Min;CHEN Chao-gen;GE Yu-fei;QI Xue-dong;LIANG Hong;TONG Zhao-qi;College of Computer and Communication Engineering in China University of Petroleum;School of Geosciences in China University of Petroleum;
Abstract:To meet the demands of mass data processing and high performance integration of resources in geophysical prospecting field, a distributed parallel processing platform based on grid technology is constructed. The architecture and key technology of the platform is described in detail. The platform, which includes two virtual organizations, was successfully deployed. Wave field forward modeling of the Marmousi model, fuzzy clustering data division strategy based pre-stack depth migration, and integral method pre-stack time migration were parallelly processed on the platform. The stability of the platform and the effectiveness of the parallel task partition strategy were proved. The efficiency was comparable with the traditional parallel platform. The platform can be applied to larger scale resources to expand the scale of parallel processing, improve resource utilization, and reduce the time of data processing.
Keywords:grid   distributed-parallel   meta-scheduling   seismic data processing
本文献已被 CNKI 等数据库收录!
点击此处可从《中国石油大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《中国石油大学学报(自然科学版)》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号