Method and device for writing data into

Abstract

The invention discloses a method and device for writing data into. The method includes obtaining of the data to be written into, a strip matched with the volume of the obtained data to be written into is determined from a disk storage system according to the volume of the obtained data to be written into, wherein the disk storage system includes multiple strips different in depth, each strip includes n data strip units and m verification strip units, the data to be written into is divided into n data strip units, the m verification strip units are calculated out through an erasure code algorithm, and the n data strip units and the m verification strip units are written into the determined strip. According to the volume of the data to be written into, the strip of the suitable depth is selected, the data to be written into is written into the selected strip through the erasure code algorithm, and the effects that the throughput is improved by selecting the strip with the large depth when the volume of the data is large, and when the volume of the data is small, the strip with small depth is selected, so that the risks of delayed writing and data losing are reduced.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of data storage, and in particular, to a data writing method and device. Background technique [0002] Erasure coding (EC) is a data protection method, which divides the original data into data blocks (strips), encodes multiple data blocks through algorithms to form one or more redundant data blocks, and divides the data into one or more redundant data blocks by continuous A stripe (stripe) composed of multiple data blocks and associated redundant blocks is stored in different storage spaces, so that data can be reconstructed after a part of the storage space fails. Erasure coding technology is a very good compromise between disk space utilization and data reliability, and has been widely used in distributed storage systems. by figure 1 For example, D0, D1, D2, D3, D4, and D5 are data blocks, and P0, P1, and P2 are redundant data blocks obtained through algorithm calculation, forming ...

AI Technical Summary

Problems solved by technology

This process has data consistency risks (especially when the disk group is distributed in a distributed storage system, such as some strips complete writing, while some strips fail to write, resulting in disordered data in the entire stripe), and consumes a lot of additional disk io performance

Images