Distributed block storage performance optimization method based on ALUA and local cache

Abstract

The invention discloses a distributed block storage performance optimization method based on ALUA and a local cache. An ALUA multi-path mode is adopted, wherein the connection priorities of the clients in the ALUA multi-path mode are divided into AO and AN; and the client distinguishes AO and AN in multiple paths, IO requests are completed through the AO path preferentially. Compared with the methods in the prior art, the distributed block storage adopts a mode of combining an ALUA multi-path mode and the local cache to improve the performance; and the distributed block storage adopts the ALUAmulti-path mode and a cluster cache equalizer to dynamically adjust the path priority and equalize the cache utilization rate of each node of the cluster. According to the method, the advantages of distributed storage and stand-alone storage are fused so that the IO path has the excellent performance of high reliability and local caching at the same time.

Description

technical field [0001] The invention relates to the technical field of data storage, in particular to a distributed block storage performance optimization method based on ALUA and local cache. Background technique [0002] For distributed block storage, the current mainstream multi-path mode between the client and server is the AA mode. The client establishes iSCSI / FC connections with multiple storage nodes to achieve high path reliability and load balancing. In the AA mode, it is necessary to ensure real-time data synchronization between multiple paths from the client to each Target, so there is no local cache under the Target service. If there is a cache, it will cause data inconsistency between paths, such as figure 1 Shown is the session connection diagram of traditional mainstream distributed block storage; for traditional stand-alone storage, there is a local cache. like figure 2 Shown is a traditional stand-alone storage architecture diagram, and the local cache i...

AI Technical Summary

Problems solved by technology

Because of the addition of the local cache layer, the write cache path is shortened, and the client perceives the write to be faster. Because the read cache is added, some reads can be directly hit and returned from the cache, the read path is shortened, and the read and write speeds are improved.

This is also one of the main reasons why the performance of distributed storage is not as good as that of stand-alone storage under the same configuration.

Images