Distributed metadata management method for distributed file system

A first terminal device, including:

• one or more processors;
• a storage device on which one or more programs are stored;
• when the one or more programs are executed by the one or more processors, the one or more processors implement operations of:
• receiving, by the server cluster, a request to access a target metadata, wherein the request is sent by a first client, the request carries an ID of the target metadata, the ID includes an area ID of the target metadata and an index ID of the target metadata;
• receiving, by the server cluster, the area ID from the first client;
• parsing, by the server cluster, the area ID and routing to a first target partition corresponding to the area ID;
• finding, by the server cluster, the target metadata in the first target partition based on the index ID;
• and
• returning, by the server cluster, the target metadata to the first client,
• wherein a root partition stores all area IDs of various partitions of the server cluster for storing metadata, and before the receiving the request to access the target metadata, the method further comprises:
• actively sending, by the server cluster, all the area IDs stored in the root partition to a first number of connected clients, wherein the root partition refers to a first partition of the server cluster for storing the metadata, the first client belongs to the first number of the connected clients, and each of the first number of connected clients stores all the area IDs in a local cache area such that the area IDs stored in the local cache area of each of the first number of connected clients are kept consistent with the area IDs stored in the root partition,
• wherein before the receiving the area ID from the first client, the method further comprises:
• searching the area IDs stored in the local cache area of the first client for the area ID of the target metadata, and
• wherein the area ID of the target metadata refers to a mark that determines the first target partition in the various partitions where the target metadata is stored and the index ID of the target metadata is a mark that determines the target metadata.

A non-transitory computer-readable storage medium on which a computer program is stored, wherein, the program when executed by the processor implements operations of:

• receiving, by the server cluster, a request to access a target metadata, wherein the request is sent by a first client, the request carries an ID of the target metadata, the ID includes an area ID of the target metadata and an index ID of the target metadata;
• receiving, by the server cluster, the area ID from the first client;
• parsing, by the server cluster, the area ID and routing to a first target partition corresponding to the area ID;
• finding, by the server cluster, the target metadata in the first target partition based on the index ID;
• and
• returning, by the server cluster, the target metadata to the first client,
• wherein a root partition stores all area IDs of various partitions of the server cluster for storing metadata, and before the receiving the request to access the target metadata, the method further comprises:
• actively sending, by the server cluster, all the area IDs stored in the root partition to a first number of connected clients, wherein the root partition refers to a first partition of the server cluster for storing the metadata, the first client belongs to the first number of the connected clients, and each of the first number of connected clients stores all the area IDs in a local cache area such that the area IDs stored in the local cache area of each of the first number of connected clients are kept consistent with the area IDs stored in the root partition,
• wherein before the receiving the area ID from the first client, the method further comprises:
• searching the area IDs stored in the local cache area of the first client for the area ID of the target metadata, and
• wherein the area ID of the target metadata refers to a mark that determines the first target partition in the various partitions where the target metadata is stored and the index ID of the target metadata is a mark that determines the target metadata.