Data processing method and device

Abstract

The invention provides a data processing method and device, and the method comprises the steps: receiving a reading request of to-be-queried data, wherein the reading request comprises a reading request serial number and a main key of the to-be-queried data; querying the primary key in a memory table; when the primary key is not queried in the memory table and the read request serial number meetsa first preset condition, querying the primary key in a row cache; and under the condition that the primary key is queried in the row cache, determining and returning a key value corresponding to theprimary key. By setting the row cache, judging the relationship between the read request and the serial number range maintained by the storage engine, querying the value corresponding to the key of the to-be-queried data in the row cache, and returning the value as the query result, the query path is effectively shortened, and the query performance is greatly improved.

Description

technical field [0001] This description relates to the field of computer technology, in particular to a data processing method. This specification also relates to a data processing device, a computing device, and a computer-readable storage medium. Background technique [0002] The log-structured merge-tree (Log-Structured Merge-Tree, Lsm-Tree) storage engine is the underlying index structure of many modern databases. The basic idea of ​​the index structure is to store the data change operation in the memory first, and when the data in the memory After the scale reaches a threshold, the data in the memory is written to the disk in batches and in an orderly manner, and the newly placed data and the old data are rolled and merged through a certain layered sorting algorithm. Wherein, in the LSM-tree index structure, data is stored in the form of key-value (Key-Value) pairs. Each key (Key) stores a corresponding value (Value), and the corresponding data value (also called a da...

AI Technical Summary

Problems solved by technology

[0003] In the traditional Lsm-Tree system, different data of the same key are usually distinguished in the form of multiple versions, which are identified by increasing version numbers. After the memory table is refreshed, there are still multiple versions of the data in the persistence layer. When querying data When searching from new to old layer by layer according to the writing order, until the desired data is found, this process may require more than one I / O, which seriously affects the speed of data query and greatly affects the user experience

[0004] In the prior art, it takes a lot of time to query data, and a simpler and more convenient method is needed for data processing

Images