Data redundancy processing method based on distributed storage system

Abstract

The invention provides a data redundancy processing method based on a distributed storage system. The distributed storage system comprises at least one control server and at least two storage servers, wherein the control server is used for controlling nodes and the storage servers are used for storing the nodes, and same copies of data are stored in at least two or more storage nodes. The data redundancy processing method comprises a processing process for reading the data and a processing process for writing the data, wherein the processing processes are not ordered; the processing process for reading the data comprises the steps of randomly selecting a storage node from all the storage servers for storing data copies and reading the storage node by the system, and if the reading fails, automatically turning to the other storage nodes to access the other copies; and the processing process for writing the data comprises the step of realizing high reliability of data writing by integrating a file data partition technique, a synchronous copying function and an asynchronous updating function. According to the data redundancy processing method based on the distributed storage system, the high reliability of the data reading on the distributed storage system is realized, and the high reliability of the data writing is realized as well.

Description

【Technical field】 [0001] The invention relates to the technical field of communication equipment, in particular to a data redundancy processing method based on a distributed storage system. 【Background technique】 [0002] A distributed storage system includes multiple storage nodes interconnected by a network, and data can be distributed and stored on different storage nodes to achieve large-capacity, high-reliability and easily expandable storage applications. [0003] The data high reliability guarantee technology commonly used in existing distributed storage systems is realized through data redundancy, that is, saving a piece of data as multiple copies, so that when the server where some copies are located fails, it can still be retrieved from other normal servers. Server access, this technology can ensure high reliability of data reading. However, in the process of data writing, as long as one of the servers with replicas fails, the writing will be interrupted, that is,...

AI Technical Summary

Problems solved by technology

However, this invention cannot be realized: in the process of data storage writing of a certain storage node, if the server of the storage node fails, the writing will be interrupted, that is, the high reliability of data writing cannot be achieved.

Images