Key value storage method based on hash table and LSM tree

Abstract

The invention discloses a key value storage method based on a hash table and an LSM tree, and belongs to the field of computer storage. The method comprises the following specific steps: S1, separately storing data and indexes by using key values; S2, packaging the key value data by using a Hash index; S3, merging the data in the ordered index in skip list merging; S4, querying the key value data by using the Hash index, and returning a query result; S5, initiating data merging for the key value data log files with the number or the size exceeding the threshold value; and S6, storing an operation process and a result of data merging. Data and indexes are separately stored by adopting the idea of key value separation, the Hash indexes are utilized to provide efficient write-in and read-out operation capability externally, a rapid range query function is realized by utilizing ordered indexes based on an LSM tree, and the problems of high read and write operation delay, large data volume, read, write, space amplification and the like of an existing key value storage system are solved.

Description

technical field [0001] The invention discloses a key-value storage method based on a hash table and an LSM tree, and relates to the technical field of computer storage. Background technique [0002] As a basic software of information technology, database plays a vital role in the information age. With the rapid development of information technology, data storage and management are also facing more and more diverse challenges. The ever-changing demand scenarios for database New requirements have been put forward in terms of storage mechanism, concurrent access, storage space utilization and usage efficiency. Traditional stand-alone databases are limited by performance bottlenecks in the era of massive data and are being replaced by distributed databases. Distributed databases dynamically meet various resource requirements in high-speed business development scenarios with good scalability and relatively low hardware investment. Provide continuous impetus for business developm...

AI Technical Summary

Problems solved by technology

BadgerDB is a high-performance key-value storage system based on the idea of ​​separating data and indexes. Its overall architecture is similar to LevelDB. Sequential lookup in BadgerDB will become a random read of the disk, which will affect its read performance.

At the same time, in order to clean up the deleted data entries in the data file, it is necessary to traverse the data entries and query the LSM tree one by one. This process consumes a lot of resources and affects the overall performance of BadgerDB

Images