Distributed storage cluster, data reading and writing method, system and related device

Abstract

Disclosed are a distributed storage cluster, a data reading and writing method, a system, an electronic device, and a computer readable storage medium. The distributed storage cluster comprises N nodes, each node running a storage service. Where N is a positive integer; N of the nodes are divided into M sub-parts, each of which stores complete write data; Where M is a positive integer greater than1. When the client end needs to write data, the distributed storage cluster divides the complete write data into a plurality of data to be written, and sends the data to each node of the sub-part respectively, so as to ensure that the nodes of the same sub-part store the complete write data together. The client only needs to wait for a sub-part of the node to complete the storage, because the same sub-part of the different nodes corresponding to the data to be written is sent in parallel, so that the waiting time of the client for data writing is reduced.

Description

technical field [0001] The present application relates to the field of storage technology, and more specifically, to a distributed storage cluster, a data reading and writing method, system, electronic device, and computer-readable storage medium. Background technique [0002] With the advent of the era of big data, the amount of data generated in the Internet field is increasing every day, resulting in the technology of distributed clusters. Through the distributed cluster technology, the services on multiple nodes can provide external services at the same time, and the downtime of one service will not affect the normal business. At the same time, nodes can be added at any time to achieve expansion. In order to ensure data security, it is necessary to ensure that multiple copies of data are stored after writing, so that normal data read and write services can still be provided when a node or hard disk fails. [0003] In the existing technology, the client writes data to a ...

AI Technical Summary

Problems solved by technology

Since this method is heavily dependent on the normal state of the main storage service, when the main storage service is abnormal, the client needs to wait for a long time until a new main storage service is generated, which brings a long failover time and cannot be applied to Time Sensitive Customer Business

Images