Granted Patent B2

US 7,721,143 · App. 11/295,344 · Granted May 18, 2010

Abstract

Claims (19)

1. A method for reducing rebuild time on a Redundant Array of Independent Disks (RAID) device, comprising steps of:

beginning or resuming a rebuild process with at least one of selecting a first stripe of said RAID device or determining a stripe of said RAID device using a last checkpoint;

enabling write-back caching on a drive being built;

reading at least one of data or parity from at least one other drive, said at least one other drive and said drive being built belonging to a same stripe of said RAID device;

exclusive ORing (XORing), when a RAID level of said RAID device is 5, said at least one of data or parity to obtain a result;

treating, when said RAID level of said RAID device is 1, said at least one of data or parity as said result;

writing said result to a second drive, said second drive being a repaired, replaced, or hot-spare drive for said drive being built;

checking whether a checkpoint needs updating; and

during said rebuild process, providing a non-rebuild write operation to the drive being built, the non-rebuild write operation requesting a forced-unit-access write, wherein the drive being built returns a status responsive to the forced-unit-access write only after all write data of the forced-unit-access write has been written to and committed to media of the drive being rebuilt.

2. The method of claim 1 , further comprising a step of inquiring, when said checkpoint needs no updating, whether a last stripe of said RAID device is reached.

3. The method of claim 2 , further comprising steps of:

selecting, when said last stripe of said RAID device is not reached, a next stripe; and

repeating said steps of reading, XORing, treating, writing, checking and inquiring until said last stripe is reached.

4. The method of claim 3 , further comprising a step of setting, when said last stripe is reached, a drive state of said RAID device to in-service and a drive group state to optimal.

5. The method of claim 4 , further comprising a step of flushing cache on said drive being built.

6. The method of claim 5 , further comprising a step of disabling said write-back caching on said drive being built.

7. The method of claim 1 , further comprising a step of flushing, when said checkpoint needs updating, cache on said drive being built.

8. The method of claim 1 , further comprising a step of updating said checkpoint.

9. The method of claim 1 , further comprising a step of inquiring whether a last stripe of said RAID device is reached.

Assignments (8)