A data processing method of a non-volatile storage medium and a computer storage medium

Abstract

A data processing method of a non-volatile storage medium and a computer storage medium, after the initialization step is completed, concurrent data insertion, data query, data deletion and data update steps are performed according to user instructions, and the concurrent smooth execution is controlled by a spin lock to guarantee. If the size of the hash table needs to be changed in the data insertion and data deletion steps, the step of adjusting the size of the hash table is triggered. After the size of the hash table is adjusted, the data insertion and data deletion steps are continued. When the user issues an instruction to end access, enter the end step. The present invention reduces cascading writes to the hash table and effectively resolves hash conflicts, and at the same time reduces the data that needs to be moved when adjusting the size of the hash table, reduces the number of times the operation of adjusting the size of the hash table is performed during frequent writing, and further Improve the life of the non-volatile storage medium, ensure data consistency and improve the performance of the hash index structure on the non-volatile storage medium.

Description

technical field [0001] The invention relates to the field of computer data storage, in particular to a data processing method of a non-volatile storage medium and a computer storage medium. Background technique [0002] At present, non-volatile memory (also called persistent memory) is gradually commercialized. Byte addressing, high density and other characteristics, non-volatile memory will become the main component of the next generation memory system. Although non-volatile memory can directly exchange data with the CPU through the memory bus using load instructions and store instructions, but due to the limited write persistence and asymmetric read-write performance of non-volatile memory, the design of Data structures face many problems. [0003] Nowadays, there have been many studies on the improvement of tree-based index structures in non-volatile memory, such as B-trees, log-structured merge trees, etc., and a small number of studies have focused on hash-based index...

AI Technical Summary

Problems solved by technology

The operation of resizing the hash table usually includes rehashing and moving the data of the old hash table to the new hash table, these operations will bring huge overhead, and also block other operations of the hash table

Although effectively resolving hash conflicts can delay the time to adjust the size of the hash table, it still cannot solve the expensive overhead of adjusting the hash table operation

[0006] Some existing hash methods move less data in the old hash table as much as possible when resizing the hash table. For example, the hierarchical hash method only needs to move one-third of the hash table size operation each time. data, thus reducing the overhead of resizing the hash table; however, the layered hash method will still frequently resize the hash table, because the newly generated hash table is only twice the size of the old hash table

[0007] On the other hand, data on non-volatile memory will cause consistency problems, that is, data loss or incomplete data when a failure occurs

Solving the problem of data consistency requires the use of expensive log operations, that is, writing old data or data to be updated into non-volatile memory first, and then performing normal write or update operations on non-volatile memory Logging operations will undoubtedly bring huge overhead and reduce its lifespan

Images