Data protection method and device

Abstract

The embodiment of the present invention provides a data protection method and device, which uses a detection device to determine the first server node where time jump occurs among at least two server nodes in the distributed storage system; sends fault indication information to the first server node, The first server node exits the distributed storage system according to the failure indication information, so as to ensure that the client will not write data into the server node where the time jump occurs, thereby reducing data loss or future data problems.

Description

technical field [0001] The embodiments of the present invention relate to communication technologies, and in particular to a data protection method and device. Background technique [0002] The distributed storage system is composed of several server nodes and clients connected to each other. The server nodes are responsible for data storage, and the clients can perform operations such as writing, reading, updating, and deleting data on the server. In a distributed storage system with time as the version, the timestamp of the stored data is used as the version number of the data, so as to ensure that the version number of the data stored later must be greater than the version number of the data stored first, and when reading the same key (key) value data, the latest data can be obtained by determining the version number based on the timestamp. [0003] However, in a time-versioned distributed storage system, if the time stamp of stored data is determined by the host time of...

AI Technical Summary

Problems solved by technology

[0003] However, in a time-versioned distributed storage system, if the time stamp of stored data is determined by the host time of the server node, the server node in the system may experience a clock jump for some reason, resulting in data loss or For future data issues, for example, when using the system's Network Time Protocol (NTP) to ensure the accuracy of each server's time, when the NTP exception does not exist and the clock jump exceeds the allowed range, the standard time is 11:00, and the clock jumps to 10:00, the data stored from 10:00 to 11:00 will be overwritten by new data, or, the standard time is 11:00, and the clock jumps to 12:00, the future time will appear data

If you continue to write data to the server node where the clock jump occurs, more data will be lost or more future data will appear

Images