File system data access method and file system

Abstract

The embodiment of the invention provides a file system data access method and a file system, the file system comprises at least one computing node and a plurality of storage nodes, and the method comprises the following steps: the computing node sends a mounting request triggered for a target logical volume to a target storage node; the computing node receives a log segment corresponding to the target logical volume and check point storage position information sent by the target storage node; the computing node reads log metadata of the log segment and the check point according to the log segment and check point storage position information so as to recover the data state of the target logical volume; and the computing node performs data access processing based on the data state of the target logical volume. When the plurality of computing nodes access the target logical volume, the plurality of computing nodes can keep good read-write consistency based on the log information of the target logical volume maintained in each storage node, and read-write can be synchronously carried out, so that read-write conflicts are avoided.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a file system data access method and a file system. Background technique [0002] Computers manage and store data through the file system. In the era of information explosion, the amount of data generated by various network services has increased exponentially. In order to meet the storage capacity of massive data, data backup, data security and other needs , the distributed file system came into being. Simply put, the distributed file system expands the traditional file system fixed in a certain location to multiple file systems in any number of locations, and consists of a large number of storage nodes. File system network. Therefore, when using the distributed file system, the user does not need to care about which storage node the data will be stored on or from which storage node the data will be obtained from, but only needs to manage and store the data as if using a l...

AI Technical Summary

Problems solved by technology

[0005] However, the commonly used NFS / CIFS network shared file system has performance and scalability bottlenecks, which are mainly reflected in: in order to support multiple clients (application clients in computing nodes) to simultaneously read and write data access consistency, usually in An independent coordination server will be deployed in the file system to coordinate all access requests by providing a complex lock competition mechanism

Therefore, any client needs to request a lock from the coordinating server before reading and writing data. When the amount of data access increases, the coordinating server is likely to cause a bottleneck, making the system less scalable

Moreover, when multiple clients need to read and write the same file data, there will be lock competition, so that the client that has not obtained the lock must wait for other clients to complete the data reading and writing before they can read and write data, which will lead to data read and write failures. The delay increases, that is, the read and write conflicts are serious

Images