The metadata management based on MongoDB for EAST experiment

Abstract

EAST experiment device uses MDSPlus database to store a large number of experimental data. The users access the data through the API provided by MDSPlus. However, due to the large amount of internal data and the lack of unified resource directory, it is difficult to browse experimental data quickly. This paper proposes a metadata management system based on MongoDB to construct resource directory for data. Firstly, the system uses C++ language to scan the entire MDSPlus database to extract and build metadata. Secondly, based on the basic document in BSON format, the system adopts the form of document nested document array, collects all the metadata of each shot in the same document, constructs single metadata using stream document, and establishes metadata database. Thirdly, encapsulates the interface of typical query and cross cut statistics, and performs performance optimization based on index. Finally, a Web project is constructed based on "Spring Boot+Bootstrap" to provide users with database resource directory navigation services. The realization of Metadata database and metadata management system helps users quickly understand the database resource overview, and supports efficient and convenient access to experimental data.