Split parity spare disk achieving method in raid subsystem

• US 6,070,249 A
• Filed: 09/22/1997
• Issued: 05/30/2000
• Est. Priority Date: 09/21/1996
• Status: Expired due to Term

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lee, Hae-Seung
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): REVAK, CHRISTOPHER A
• Application Number:
  US08/935,201
• Time in Patent Office: 981 Days
• Field of Search: 714/6, 714/7, 714/766, 714/805, 711/114
• US Class Current: 714/6.21
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems

Abstract

A split parity spare disk achieving method for improving the defect endurance and performance of a RAID subsystem which distributively stores data in a disk array consisting of a plurality of disk drives and carries out an input/output operation in parallel includes the steps of: constructing the disk array with at least two data disk drives for storing data, a spare disk drive used when a disk drive fails and a parity disk drive for storing parity data; and splitting the parity data of the parity disk drive and storing the split data in the parity disk drive and the spare disk drive.

Claims

1. A split parity spare disk achieving method for improving the defect endurance and performance of a RAID (Redundant Arrays of Inexpensive Disks) subsystem which distributively stores data in a disk array consisting of a plurality of disk drives and carries out an input/output operation in parallel, said method comprising the steps of:

constructing said disk array with at least two data disk drives for storing data, a spare disk drive used when a disk drive fails and a parity disk drive for storing parity data; and

splitting said parity data of said parity disk drive, and storing the split data in said parity disk drive and said spare disk drive.

2. The method as claimed in claim 1, wherein said disk array comprises four data disk drives, one parity disk drive and one spare disk drive.

3. A split parity spare disk achieving method for improving the defect endurance and performance of a RAID (Redundant Arrays of Inexpensive Disks) subsystem which distributively stores data in a disk array consisting of a parity disk drive, a spare disk drive and a plurality of data disk drives and carries out an input/output operation in parallel, said method comprising the steps of:

splitting parity data stored in said parity disk drive into upper block parity data and lower block parity data, and storing said upper block parity data in a upper block storage region of said parity disk drive and storing said lower block parity data in a upper block storage region of said spare disk drive;

defining the prescribed number of disk drives except said spare disk drive as small parity groups, and generating small group parity data by using disk drives of a small parity group which includes said parity disk drive; and

splitting said small group parity data into small group upper block parity data and small group lower block parity data, and storing said small group upper block parity data in a lower block storage region of said spare disk drive and storing said small group lower block parity data in a lower block storage region of said parity disk drive.

4. The method as claimed in claim 3, wherein said small parity group comprises:

a first parity group consisting of a plurality of data disk drives which does not include said parity disk drive; and

a second parity group consisting of a plurality of data disk drives and including said parity disk drive.

5. The method as claimed in claim 4, wherein each of said disk drives are divided into an upper block storage region and a lower block storage region on the basis of an intermediate cylinder value of each disk drive.

6. The method as set forth in claim 5, further comprising the steps of:

detecting a failed drive;

determining whether said failed drive is any one or more of said data disk drives, said parity disk drive or said spare disk drive.

7. The method as claimed in claim 6, further comprising the steps of:

copying the lower block parity data stored in the upper block storage region of said spare drive to the lower block storage region of said parity drive when it is determined that one of said data disk drives failed;

recovering the upper and lower block data of said failed data disk drive by exclusively ORing the upper and lower block data of the data disk drives of said disk array which did not fail with the upper and lower block parity data, respectively; and

writing the recovered upper and lower block data in the upper and lower block storage regions of said spare disk drive.

8. The method as set forth in claim 6, further comprising the steps of:

exchanging the lower block parity data stored in the upper block storage region of said spare disk drive and the small group lower block parity data stored in the lower block storage region of said parity disk drive for each other, when it is determined that one of said data disk drives in each of said first and second parity groups failed;

recovering the upper and lower block data of the failed data disk drive within said second parity group by exclusively ORing the upper and lower block data of the data disk drive which did not fail within said second parity group, the upper and lower block data of said parity disk drive and the upper and lower block data of said spare disk drive;

writing the recovered upper and lower block data in the upper and lower block storage regions of said spare disk drive;

recovering the upper and lower block data of the failed data disk drive in said first parity group by exclusively ORing the upper and lower block data of each of the data disk drives which did not fail, the upper and lower block data of said parity disk drive and said recovered upper and lower block data stored in the upper and lower block storage regions of said spare disk drive; and

writing the recovered upper and lower block data of the failed data disk drive which is in said first parity group in the upper and lower block storage regions of said parity disk drive.

9. The method as set forth in claim 6, further comprising the steps of:

copying the lower block parity data stored in the upper block storage region of said spare disk drive to the lower block storage region of said spare disk drive, when it is determined that said parity disk drive failed;

recovering the upper block parity data of said parity drive by exclusively ORing the upper block data of each of said data disk drives; and

writing the recovered upper block parity data into the upper block storage region of said spare disk drive.

10. The method as set forth in claim 6, further comprising the steps of:

recovering the lower block parity data stored in the upper block storage region of said spare disk drive by exclusively ORing the lower block data of each of said data disk drives, when it is determined that said spare disk drive failed; and

writing the recovered the lower block parity data into the lower block storage region of said parity disk drive.