首页 | 本学科首页   官方微博 | 高级检索  
     

地震并行处理模式与应用框架
引用本文:王宏琳,GAO Hui-sheng.地震并行处理模式与应用框架[J].计算机学报,2001,24(2):202-208.
作者姓名:王宏琳  GAO Hui-sheng
作者单位:1. 石油地球物理勘探局
2. Bureau of Geophysical Prospecting,
基金项目:国家“八六三”高技术研究发展计划,中国石油天然气集团公司科技计划项目
摘    要:文中研究石油地震数据处理的并行计算设计模式(流水、扇出/扇入、主从和混合)和应用框架,框架和模式的目的都是复用成功的软件设计自力更生,框架可以看作一类设计模式的具体实现。针对地震数据处理模式设计和实现了GRISYS地震数据处理应用框架。利用这个框架,以往大量的串行地震处理模块,不需要任何改动,可以在工作站集群计算机或大规模并行计算机上实现并行计算,在曙光2000-Ⅱ并行计算机上试验,获得了非常高的并行处理加速比。

关 键 词:并行处理  地震数据处理  地震勘探  石油勘探  并行计算机
修稿时间:2000年1月10日

The Seismic Processing Patterns and Application Framework
WANG Hong-lin,GAO Hui-sheng.The Seismic Processing Patterns and Application Framework[J].Chinese Journal of Computers,2001,24(2):202-208.
Authors:WANG Hong-lin  GAO Hui-sheng
Abstract:The Paper describes design patterns(namely pipeline, fan in/fan out , master slave and hybrid pattern)and application framework for seismic data parallel processing. Frameworks and patterns both facilitate reuse by capturing successful software development strategies. Frameworks can be viewed as a concrete realization of families of design patterns that are targeted for a particular application domain. The GRISYS application framework has been designed and implemented based on these seismic processing patterns. With the use of this application framework, we are able to achieve parallel data processing on workstation clusters or massive parallel computers, without any revision of multitudinous serial processing modules. Parallel processing splits this type of processing modules over many processors to cut down overall computing time. It extended specific geophysical language by providing parallel statements. GRISYS can perform 2D and 3D imaging of seismic data to produce images of the subsurface. The job examples of 3D data processing have been benchmarked on Dawning 2000 II computer to examine the performance of the GRISYS parallel framework.
Keywords:parallel computation  parallel processing  application framework  seismic processing  
本文献已被 CNKI 维普 万方数据 等数据库收录!
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