Column-oriented database management system

Abstract

The embodiment of the invention discloses a column-oriented database management system. The system comprises a user interface module, an application program interface module, a core service module anda data processing module, wherein the core service module comprises a memory pool and a working thread, and the memory pool comprises M memory blocks; wherein M is greater than or equal to 2; the data processing module is used for organizing, aggregating and storing the data by taking a column as a unit and querying the data in a full-index manner; wherein a data structure adopted by the data processing module is a single-list data structure. According to the embodiment of the invention, a user does not need to compile complex query statements, and the storage of unstructured data and structured data, the storage of relational and non-relational data and the combined query under any condition are supported.

Description

technical field [0001] The invention relates to the technical field of databases, in particular to a column-oriented database management system. Background technique [0002] Currently, data storage management solutions for big data systems include traditional relational database RDBMS and non-relational database NoSQL. Because the massive data on the network has no fixed structure, and there are complex and diverse relationships among them, data engineers often need to use multiple relational databases and NoSQL databases at the same time to fulfill their own data processing needs. This multi-type mixed use method makes the program logic complex, the process is cumbersome, the execution is slow, the error rate is high, the scalability is weak, the ease of use is poor, and the transplantation is difficult, especially bringing additional inconvenience to big data statistical analysis and data mining. [0003] The bottlenecks of traditional relational databases are mainly man...

AI Technical Summary

Problems solved by technology

This multi-type mixed use method makes the program logic complex, the process is cumbersome, the execution is slow, the error rate is high, the scalability is weak, the ease of use is poor, and the transplantation is difficult, especially bringing extra inconvenience to big data statistical analysis and data mining.

[0003] The bottlenecks of traditional relational databases are mainly manifested in: single data type, low performance of high concurrent read and write, limited support capacity, low scalability and availability of database, high construction and operation and maintenance costs, etc.

[0004] The non-relational database NoSQL has greatly improved the access efficiency of the database, but it has brought more serious problems: the database is basically not universal, the transplantation ability is poor, and the data migration is difficult; the weakening of the data relationship makes the data statistical mining and analysis difficult; The format needs to be customized by the user, and the management of the data format is handed over to the upper application, which increases the system complexity of developing software

Images