To put it simply, distributed storage is to store data dispersedly on multiple storage servers, and form a virtual storage device with these dispersed storage resources. In fact, data is stored dispersedly in various parts of the enterprise. corner.
Distributed storage simply means to store data dispersedly on multiple storage servers and form a virtual storage with these dispersed storage resources. Devices, in fact, data are stored dispersedly in every corner of the enterprise.
For example:
If you want to store a 10PB video file, it cannot be stored on your own computer. It needs to be stored on multiple servers. Each server is a datanode. When you need When obtaining files, accessing them one by one is too inefficient. You can find an intermediary to store the information on which server the file is located. The file can be quickly found through the information. This middleman is the namenode, and the information stored about the file is metadata. For data security, a copy of each small video (storage unit) is stored on a different server. When a server stops running, you can find the data you want by accessing the backup. This storage method is called distributed storage. .
The distributed storage architecture consists of three parts: Client, metadata server and data server.
The client is responsible for sending read and write requests, caching file metadata and file data.
The metadata server is responsible for managing metadata and processing client requests. It is the core component of the entire system.
The data server is responsible for storing file data and ensuring data availability and integrity. The advantage of this architecture is that performance and capacity can be expanded at the same time, and the system scale is highly scalable.
The above is the detailed content of What is the principle of distributed storage?. For more information, please follow other related articles on the PHP Chinese website!