Data consistency concurrent maintenance method

Abstract

The invention discloses a data consistency concurrent maintenance method. The method comprises the steps that 1, a node broadcasts a request for other nodes in a network for a synchronous block in a current synchronous period and calculates a synchronous anchor of the current synchronous period, and if the expected synchronous block is not obtained, the node calculates the synchronous anchor of the stored synchronous block or initializes a starting synchronous block S0 and the synchronous anchor thereof; 2, a node I with the data addition requirement writes the synchronous anchor of the current synchronous period into data A0, and an array [A]N is generated as needed; the data A0 is to be added into consistency data in the synchronous period; 3, the nodes concurrently complete construction and broadcasting of the array [A]N of the nodes; 4, the nodes filter all received [A]N with the identical synchronous anchor, abnormal data is removed, and the synchronous block of the next synchronous period is constructed. Data of the synchronous blocks can be concurrently generated, and meanwhile the whole data structure can still keep the whole consistency on the dependency.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to a data verification method which has global consistency and is friendly to parallel processing processes in a network distributed computing environment. Background technique [0002] Network Distributed Computing [0003] Computing units (including but not limited to computers) in a network environment operate independently, and cooperate to complete specific computing tasks through specific communication protocols and application software. When distributed computing needs to rely on the same data set, all computing nodes rely on a centralized storage management mechanism to read and write the required data. This method of reading and writing data is applicable under fully managed computing conditions. In the absence of centralized storage management, although the existing data verification methods can solve the verification and error correction in transmission, be...

AI Technical Summary

Problems solved by technology

[0014] To sum up, the existing solutions for multi-node distributed computing rely on centralized storage and serialized processing to solve data consistency problems. In a decentralized scenario, if you want to maintain data consistency, you must rely on data Serialization operation, the serialization method will block the operations of other nodes, even nodes without context dependencies often have to wait for the operations of other nodes to complete, which greatly reduces the operation efficiency

However, the post-processing method to solve the problem of data consistency cannot meet the data requirements of distributed systems in real-time operation.

Images