Virtual experiment-oriented unstructured data accessing method

Abstract

The invention discloses a virtual experiment-oriented unstructured data accessing method, which comprises the following steps of: (1) taking all files corresponding to a given experimental component as an experimental device object, and taking the object as a minimum data accessing unit so as to avoid hard disk input / output (IO) caused by the accessing of a great number of small files and improve throughput; (2) performing data updating operation in a relatively lower ratio by adopting a message queue, and asynchronously performing data modification operation so as to decouple application from data and simultaneously solve the problem of combination in the concurrent writing of the data; (3) dynamically binding corresponding schemas in data resolution, wherein experimental files have different types and formats and each type of data corresponds to a resolution schema; (4) loading all the data defined in a subject or specific experiment of frequently accessed data in advance according to the subject correlation and specific experiment correlation of the data in a virtual experiment, and simultaneously adopting a concatenated caching structure of which caching sizes are gradually increased from top down and an upper layer is a subset of cached data in a lower layer; and (5) for achieving independence from a platform and a language, providing a plurality of types of RESTful interfaces according to the data needs of a user by adopting a hyper text transfer protocol (HTTP)-based REST architecture, and providing a uniform resource identifier (URI)-based safety certification mechanism and a complex query method.

Description

technical field [0001] The invention belongs to the field of computer applications, and in particular relates to a virtual experiment-oriented unstructured data access method, which is suitable for complex queries of massive unstructured data in virtual experiments, and provides simple, unified, safe and efficient access interface. Background technique [0002] With the improvement of informatization, data has exceeded its original scope. It includes various business operation data, report statistics, office documents, emails, hypertext, tables, reports, pictures, audio and video, etc. . People use massive data to describe huge, unprecedented, and ever-growing data. According to industry analysis reports, unstructured data accounts for more than 85% of the entire information volume, and the data volume is extremely huge, which is the core of information resource management. [0003] As an effective supplement to traditional experimental teaching, virtual experimental teac...

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to provide a virtual experiment-oriented unstructured data access method, which can establish indexes for massive unstructured data in various forms, quickly locate and search, and provide a unified access interface, so as to ensure the accuracy and accuracy of virtual experiments. Stable and efficient operation provides a storage support environment, and provides complex query methods to meet the needs of different levels

Images