Data synchronization method and device for distributed storage system

Abstract

The invention discloses a data synchronization method and device for a distributed storage system. The method comprises the following steps: respectively generating and storing service statistical data of corresponding buckets of a first data center and a second data center; sending a data synchronization request to a second data center, and sending the service statistical data of the corresponding bucket of the first data center to the second data center; determining a data synchronization rate according to the received service statistical data of the corresponding bucket of the first data center and the stored service statistical data of the corresponding bucket of the second data center; and performing data synchronization between the first data center and the second data center based on the data synchronization rate. According to the invention, the appropriate speed of data synchronization can be determined in a targeted manner based on respective business conditions of different data centers, so that maximum data balance is provided for different data centers, and the potential loss of business data is reduced while the data synchronization speed is improved.

Description

technical field [0001] The present invention relates to the field of resource allocation, and more specifically, to a data synchronization method and device for a distributed storage system. Background technique [0002] With the explosive growth of massive unstructured data, distributed object storage has become the storage cornerstone for users to build data center architectures, and more and more key businesses are connected to distributed object storage. Distributed object storage products not only need to ensure the storage needs of customers for massive data, but also meet the disaster recovery backup of key data. For example: in the financial industry, there are multiple data centers in one city, and how multiple data centers synchronize, back up, and share data; when a cluster fails, it is necessary to ensure that the user business continues to run. When different data centers perform data synchronization, the synchronization speed of customer business idle scenario...

AI Technical Summary

Problems solved by technology

When different data centers perform data synchronization, the synchronization speed of customer business idle scenarios is often relatively slow, resulting in out-of-synchronization of data in different data centers for a long time; in busy customer business scenarios, data synchronization is also related to user read and write services. Due to competition, the front-end business is affected by data synchronization, and the balance between different data center businesses and synchronous data cannot be achieved

[0003] There is currently no effective solution to the problem that data synchronization between different data centers is difficult to balance in the prior art

Images