Distributed data storage method and device based on data feature pooling

Abstract

The invention relates to a data storage method and device, belonging to the field of data storage, in particular to a distributed data storage method and device based on data feature pooling. Including: the feature division step, pre-planning the fault domain algorithm according to the characteristics of business data, the algorithm will automatically divide the data into different feature types; the strategy formulation step, formulating distributed storage data optimization strategies for different feature types; the strategy loading step, The storage device loads the corresponding data storage optimization strategy according to the data feature type of the sub-pool; in the optimization execution step, the object storage device executes the optimization operation according to the relevant actions defined by the data storage optimization strategy. Therefore, the present invention has the following advantages: the forward write performance of the node disk where the sequential feature pool is located is basically at full bandwidth, and the disk where the random feature pool is located also basically reaches the random performance of the limit, which significantly improves the data operation efficiency of distributed storage .

Description

technical field [0001] The invention relates to a data storage method and device, belonging to the field of data storage, in particular to a distributed data storage method and device based on data feature pooling. Background technique [0002] Distributed storage is a data storage technology. In simple terms, it is to disperse and store data on multiple storage servers, and form these dispersed storage resources into a virtual storage device. In fact, data is distributed and stored in each server. corner. [0003] Traditional network storage systems use centralized storage servers to store all data, and storage servers become the bottleneck of system performance and the focus of reliability and security. The distributed storage system uses multiple storage servers to share the storage load, and uses the location server to locate and store information, which effectively improves the reliability, availability and access efficiency of the system. [0004] At present, the dis...

AI Technical Summary

Problems solved by technology

Although the logical pool splitting scheme is relatively simple, this scheme is unable to adapt to the performance requirements of the business and cannot perceive the data characteristics of the business, which means that this scheme cannot make the most reasonable optimization for different businesses. There are certain requirements, but the requirements are not very high. According to this pool division scheme, the business must be put into a high-performance pool, which increases the construction cost of the overall storage

Images