气象大数据存储体系设计与实现  

Design and implementation of meteorological big data storage system

作　　者：徐拥军[1] 何文春[1] 刘媛媛[1] 王琦[1] Xu Yongjun;He Wenchun;Liu Yuanyuan;Wang Qi(National Meteorological Information Center,Beijing 100081,China)

机构地区：[1]国家气象信息中心,北京100081

出　　处：《电子测量技术》2020年第22期19-25,共7页Electronic Measurement Technology

摘　　要：为满足天气预报、气候预测、决策服务和公众服务等气象业务对海量气象数据存储与服务的需求,实现气象业务集约化的目标。基于业务数据应用特征,结合多种分布式存储技术,设计了一套气象大数据存储体系。通过规范数据存储结构,利用数据同步、分级存储以及数据服务接口等方面的关键技术,解决了气象数据在分布式异构存储技术间的统一管理和服务的问题。经过应用实践证明,气象大数据存储系统存储能力达到了30 PB,数据分级管理流程高效,数据统一服务响应时效在1 s以内,具备直接支撑气象业务的能力。In order to solve the storage of meteorological big data, and to meet the needs of weather forecasting, climate prediction, decision-making services and public services for data services. A set of meteorological big data storage system is designed combined with application characteristics and based on a variety of distributed storage technologies, by standardizing the data storage structure, using the key technologies of data synchronization, hierarchical storage and data service interface. Through the practical verification, it is proved that the system can store 30 PB data, the hierarchical management process of data is efficient, data services is less than 1 s. So, the meteorological big data storage system is feasible, which not only improves the storage capacity of meteorological big data, but also meets the needs of application services.

关 键 词：分布式存储 气象大数据 数据同步 分级存储 数据统一服务 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]