Edge storage database based on improved LSM tree

Abstract

The invention discloses an edge storage database based on an improved LSM tree, and the database comprises: an index module which is responsible for the maintenance of data internal keys and address information; a value storage module which is responsible for maintaining data internal keys and actual value information; a database function module which is used for converting various complex data structures into internal key-value pairs; a network module which is used for providing a multiplexed TCP service; a database synchronization module which is used for providing master-slave synchronization and multi-master synchronization among multiple nodes of the database; and a user service module which directly interacts with a user through various standardized interfaces. Through the improvement of the LSM tree and the cooperation of multiple modules, a high-performance, low-loss, high-availability and standardized edge storage database is successfully constructed, and compared with an existing Internet-of-Things edge data storage system, the write-in amplification of the system is remarkably reduced.

Description

technical field [0001] The invention relates to the technical fields of edge storage, Internet of Things, embedded, flash storage, and database, in particular to an edge storage database based on an improved LSM tree. Background technique [0002] The rapid development of IoT technology has spawned a large number of terminals that are physically located at the edge of the network, forming an intricate edge network. But at the same time, due to network conditions, security and privacy, cost budget and other reasons, traditional centralized cloud computing is powerless in the face of edge networks. Therefore, edge computing, which is closer to the data source, lower cost, less traffic, and more secure, has developed rapidly, and edge storage, as the core link in edge computing, is becoming more and more important. [0003] Edge storage differs from traditional data storage in two significant ways: [0004] One is the difference in application scenarios: first, the data store...

AI Technical Summary

Problems solved by technology

More seriously, flash memory devices such as SD cards, similar to solid-state drives, are highly sensitive to write amplification. Excessive writing will lead to a significant reduction in the life of SD cards. Therefore, IoT flash memory devices require low write amplification storage solutions.

[0006] In addition, there is no uniform standard in the field of edge computing at present, and the widely used solutions are mainly local databases such as SQLite, which lack the data synchronization capability required for edge data integration and increase development costs.

Images