Disk array system supporting grouping-free overall situation hot standby disks based on flexible redundant array of independent disks (RAID)

Abstract

The invention relates to a disk array system supporting grouping-free overall situation hot standby disks based on flexible redundant array of independent disks (RAID). The disk array system comprises one or more RAID, one or more free disks, 0 or more than one overall situation hot standby disks, a system disk and a memory operating system. The RAID are established by using '-C' or '-create' parameters of mdadm software in a Linus operating system and are managed by mdadm. The system disk is connected with each of RAID, the overall situation hot standby disks and the free disks. Contents memorized by the system disk include but not limited to the memory operating system and application programs managing the RAID. The memory operating system is memorized in the system disk, is the Linus operating system with an expanded kernel and comprises a man-machine interaction module, an overall situation hot standby disk establishing module, an overall situation hot standby disk retrieval module, an overall situation hot standby disk deleting module and an RAID damaged disk detecting and processing module. The overall situation hot standby disks are not limited by RAID groups, do not need configuration files and facilitate data transfer.

Description

technical field [0001] The invention relates to a disk array system supporting non-grouped global hot spare disks based on soft RAID, and belongs to the field of computer data storage. Background technique [0002] In the field of large-capacity data storage, various types of RAID are usually used to build infrastructure for data storage. RAID is the abbreviation of Redundant Arrays of Independent Disks, which is called redundant array of independent disks, or disk array for short. Commonly used RAIDs include RAID0, RAID1, RAID5, and RAID6. Among them, RAID0 is mainly to improve read and write performance, and does not have redundancy capability. Other RAIDs have different degrees of data redundancy: RAID1 mirrors the disks, and the redundancy reaches 50%. When a disk is damaged, another mirror disk can be used to rebuild it; each stripe of RAID5 contains 1 parity block, support arbitrary damage to one of the disks, and rebuild data through parity blocks on other disks; ea...

AI Technical Summary

Problems solved by technology

In a soft RAID environment, from the perspective of implementation, the set global hot spare disk needs to record the information of the group to which the RAID belongs in the "mdadm.conf" configuration file. Hot spare disk management, there are the following problems: ① "mdadm.conf" configuration file belongs to the configuration file of the operating system, in the overall migration process of the disk, the configuration file will not migrate along with it, resulting in Application reliability is low, and it is not conducive to data migration

②When there are multiple RAID groups in the disk array system based on soft RAID, a global hot spare disk can only be responsible for the RAID member disks in the group it is responsible for, and cannot realize the sharing of global hot spare disks among different RAID groups

Images