摘要
油气勘探数据作为大数据的种类之一,具备大数据所有数据量(Volume)、时效性(Velocity)、多样性(Variety)、可疑性(Veracity)属性。属性分析是大数据处理的必要前置步骤。现有高密度、宽方位地震所采集数据量巨大,在地震资料处理环节及地震数据属性分析环节出现重大时效问题,需要计算机硬件设备扩充与软件实现效率和方式的改变。文中通过例证资料处理过程中地震数据属性分析流程,判断分析的效率和分析方式对资料处理周期和品质的影响。通过分析地震资料处理软硬件特点,提出多线程共享内存及优化数据I/O粒度的数据分析等方法提高效率,测试应用效果表明该方法大幅度提高海量地震数据属性分析效率。
Oil and gas field exploration data are big data, which have Volume, Velocity, Variety, Veracity attributes of big data, data attribute analysis is a necessary pre steps. With the development of seismic exploration instruments and seismic exploration technique, seismic data become bigger. Especially the seismic data with the high density wide azimuth attributes, is easily reach tens or even hundreds of Terabytes. It brings great challenge to the seismic data processing and attribute analysis. The chaUenge needs to upgrade computer hardware and improve software efficiency. Seismic attribute analysis is indispensable in the data processing, its efficiency and analysis mode is the key problem to data processing cycle and quality. After analyzing the existing seismic data processing characteristic, through the use of multi-thread shared memory and optimization I/O of data granularity can improve efficiency. The test result shows that the method can highly improve the mass seismic data attribute analysis efficiency.
出处
《信息技术》
2015年第11期197-200,204,共5页
Information Technology
关键词
计算机应用
大数据
属性分析
多线程
共享内存
数据粒度
computer applications
big data
attribute analysis
multi-thread
shared memory
data granularity
