Cluster distributed type lock management method based on storage area network SAN

Abstract

A distributed lock management method based on a storage area network SAN belongs to the network storage field. The distributed lock management method is characterized in that all node servers of a cluster form multiple master-slave mode sub-clusters, namely, Quorum; one node in the sub-clusters is used as the master server and the other nodes are slave servers; the multiple sub-clusters include one State Quorum and a plurality of Lock Quorums, which are respectively in charge of the management of the birth-death state of the system nodes and the management of a read-write visit lock in a name space; the State Quorum is formed by all the nodes and the Lock Quorum is formed by nodes of arbitrary number, the sizes and the structures of the State Quorum and the Lock Quorum can be flexibly configured by the client; both the State Quorum and the Lock Quorum follow the sub-cluster detection protocol, and all the nodes can select the only master server in the Quorum through the protocol to configure a normal Quorum; all the Lock Quorums respectively manages the different sections of the name space, are not overlapped and are respectively vested with different weights according to different service capacities; all the Lock Quorums independently count the load conditions, therefore when the lock management task charged by one Lock Quorum is too heavy, the load balancing process is triggered to distribute part task of the Lock Quorum to the other Lock Quorums. The invention is easy for flexible configuration and has good properties, high extensibility and high reliability.

Description

technical field [0001] The cluster distributed lock management method based on the storage area network SAN belongs to the field of network storage systems. Lock management is very important for maintaining system consistency and ensuring the correctness of concurrent access. Load scheduling in the network, cluster global data snapshot backup and other fields. Background technique [0002] The distributed lock management system is used to control concurrent access to shared resources, improve the efficiency of concurrent access, and at the same time ensure the atomicity of modification operations on shared resources, thus ensuring the consistent state of the system. The traditional lock management system relies on the central lock management server to provide access control to the entire data resource name space. In this solution, the central lock management server is likely to become a system bottleneck, and the scalability of the entire system is poor. At the same time, th...

AI Technical Summary

Problems solved by technology

The traditional lock management system relies on the central lock management server to provide access control to the entire data resource name space. In this solution, the central lock management server is likely to become the bottleneck of the system, and the scalability of the entire system is poor. At the same time, the central lock management server The single node failure problem of the server will also affect the reliability of the system; the distributed lock management system is adopted in the object-oriented distributed file system represented by the cluster parallel file system Lustre, and the access control distribution of the OSD name space of the object storage server is distributed. Into several metadata management servers MDS, this greatly reduces the burden on a single server, but in this solution, the distribution of namespaces in the metadata management server MDS is static, lacks flexibility, and cannot adapt to the dynamic load At the same time, this type of lock management system adopts a hot standby server solution, which improves system reliability to a certain extent; the cluster system represented by the global file system Gloable FS adopts a lock management system based on sub-cluster Quorum, The advantage is that the reliability of the system is greatly improved. All servers in Quorum are kept strictly synchronized through the lock update protocol inside Quorum. When one or more of them fails, other servers can still form a new Quorum to continue to provide services according to the agreement. However, in this system, there is only one node inside the Quorum, called the main server Master, which can really provide external services, so there is a performance bottleneck

Images