RS code node repairing method based on genetic algorithm under background of multiple data centers

Abstract

The invention discloses an RS code node repairing method based on a genetic algorithm under the background of multiple data centers. The invention aims to provide an optimal bottleneck bandwidth path selection method based on a genetic algorithm aiming at the problem that a global optimal bottleneck bandwidth repair scheme cannot be obtained under the background of multiple data centers in a traditional data repair mode, a repair tree with the maximum bottleneck bandwidth is generated according to the computing power of nodes and the bandwidth between the nodes, and the network bandwidth consumption and the repair time generated during node repair are effectively reduced. According to the RS code node repairing method based on the genetic algorithm under the background of the multiple data centers, the problems that a traditional star repairing scheme and an assembly line repairing scheme are large in repairing time delay and a traditional tree repairing scheme is large in bandwidth consumption are solved, redundant data transmission is reduced, the repairing efficiency is improved, and the repairing time is shortened.

Description

technical field [0001] The invention belongs to the field of distributed erasure code storage systems, and in particular relates to an RS code node repair method based on a genetic algorithm under the background of multiple data centers. Background technique [0002] Due to its excellent performance and low construction cost, the distributed storage system has become the mainstream storage system in the field of large-scale data storage. However, because the underlying devices of the distributed storage system generally use cheap commercial hardware, node failure has become a normal state. . [0003] In order to prevent business loss caused by data failure caused by node failures, distributed storage systems often use erasure codes and multi-copy redundant data to ensure data integrity and reliability. Erasure codes are less expensive because of additional storage It is widely used, but when some nodes fail and the data becomes invalid, the erasure code needs to read other ...

AI Technical Summary

Problems solved by technology

[0002] Due to its excellent performance and low construction cost, the distributed storage system has become the mainstream storage system in the field of large-scale data storage. However, because the underlying devices of the distributed storage system generally use cheap commercial hardware, node failure has become a normal state.

[0003] In order to prevent business loss caused by data failure caused by node failures, distributed storage systems often use erasure codes and multi-copy redundant data to ensure data integrity and reliability. Erasure codes are less expensive because of additional storage It is widely used, but when some nodes fail and the data becomes invalid, the erasure code needs to read other data blocks and perform encoding and decoding to restore the invalid data. This process will generate a lot of repair traffic overhead, and the repair speed is slow

Images