Software and hardware cooperative multi-controller disk array designing method

Abstract

The invention provides a software and hardware cooperative multi-controller disk array designing method. Along with increasing of controller quantities, heartbeat calculating load of each controller is fixed, and fault judging time is also fixed. Bidirectional annular heartbeats can enable failures of a plurality of continuous controllers to be faster found, and have good expansibility. The bidirectional annular heartbeats are operated on a system software layer and belong to a software heartbeat mechanism. Due to the fact that the software heartbeats are operated in the controllers, when the controllers fail, the split-brain problem is easy to cause. Therefore, in design of the software and hardware cooperative multi-controller disk array designing method, the software heartbeat mechanism is cooperated with a hardware arbitration mechanism, the split-brain problem is avoided, state monitoring reliability of the controllers is improved, delay of finding a fault is reduced, efficient and reliable system state maintenance is achieved, and normal and stable operation of a storage service is ensured.

Description

technical field [0001] The invention relates to the field of computer systems and storage, in particular to a multi-controller disk array design method with software and hardware coordination. Background technique [0002] With the advancement of technology, the performance of disk arrays is continuously enhanced, and their functions are becoming more and more perfect, and their status in IT facilities is becoming more and more important. The stable and reliable operation of storage devices is of increasing significance to data security and business production. The structure of the disk array has gradually developed from the initial single control to dual control and multi-control. The dual-controller disk array has two mutually standby controllers to work in coordination, and other components also adopt redundant design. Compared with the single-controller disk array, the reliability and performance are greatly improved, and it can meet the requirements of common business a...

AI Technical Summary

Problems solved by technology

Generally speaking, there may be a delay of up to 10 seconds from the occurrence of the fault to the judgment of the fault, thus blocking the user's data read and write requests for about 10 seconds, which is intolerable for critical applications

[0007] 2) It is prone to split-brain problems, resulting in wrong switching of resources, thus affecting the stable operation of the application

Split-brain is mainly caused by the inability of each controller to accurately judge the status of the peer controller; when a split-brain occurs, the heartbeat detection between controllers fails, misjudging that the peer controller has failed, and thus taking over the operation of the peer controller services and writes to storage volumes, resulting in data corruption and service interruption

In a multi-controller disk array, a split-brain that occurs between two controllers will spread to all controllers, resulting in the interruption of all services and data corruption.

Images