Cache cluster-based cache method and system

Abstract

The invention provides a cache cluster-based cache method. The method comprises the following steps: determining a subarea number of a cache subarea corresponding to to-be-written data by a cache client; querying information of a main node of the cache subarea corresponding to the to-be-written data from a view service node according to the subarea number; receiving the information, returned by the view service node, of the main node of the cache subarea corresponding to the to-be-written data, and sending a write request to the main node of the cache subarea corresponding to the to-be-written data according to the main node information, wherein the write request carries the to-be-written data; writing the to-be-written data into a local write-back cache by the main node of the cache subarea corresponding to the to-be-written data according to the write request; and acquiring information of standby nodes of the cache subarea corresponding to the to-be-written data from the view service node, and copying the to-be-written data into the standby nodes of the cache subarea corresponding to the to-be-written data. The method is capable of enhancing the reliability of cache-scale clusters.

Description

technical field [0001] The invention relates to the field of storage, in particular to a cache method and system based on cache clusters. Background technique [0002] In existing data centers and storage products, high-speed non-volatile cache devices such as flash memory (flash) are widely used to accelerate the access speed of background shared storage. There are two ways to write the cache: write-through and write-back. [0003] Write-through refers to writing a copy to the shared storage device while writing to the cache device. The advantage is that the operation is simple, the data in the cache device and the shared storage device are consistent, and it is not easy to lose data; the disadvantage is that the data access speed is slow, and the overall read and write speed is greatly affected by the shared storage device, which improves the input per second of the system The number of outputs (input output per second, IOPS) is more difficult. [0004] Write-back (writ...

AI Technical Summary

Problems solved by technology

[0007] It can be seen from the above that the method of dual write-back cache devices mirroring each other can solve the problem of consistency between the two copies of the write-back cache to a certain extent, but this method only supports two write-back cache devices mirroring each other, which cannot meet large-scale Cluster Reliability Requirements

Images