摘要
利用数据网格技术实现海量地震资料并行处理需要解决动态、自治、异构的高性能计算系统之间的分布数据管理问题。根据石油领域资源现状,将该领域高性能资源划分为不同的虚拟社区,并建立分布数据管理体系结构,详细描述元数据模型与副本定位算法等关键技术。实现并部署了基于Web portal的地震资料数据管理平台,系统运行稳定、可操作性强、可扩展性好。
Parallel processing of massive seismic data with data grid technology has to resolve distribute data management problem between dynamic,autonomous,heterogeneous high-performance computing systems.The high performance resources in petroleum industry are divided into different virtual communities according to the status quo of the resource characteristics.Distributed data management architecture is established as well.The key technologies including metadata model and replica location algorithm are expatiated.The Web portal-based seismic data management platform is implemented and deployed,the system runs stably,and has good operability and scalability.
出处
《计算机应用与软件》
CSCD
北大核心
2013年第2期248-252,共5页
Computer Applications and Software
基金
中国石油天然气集团公司中青年创新基金项目(07E1024)
关键词
数据网格
分布数据管理
元数据模型
副本定位
Data grid Distributed data management Metadata model Replica location
