Adaptive rebuild scheduling scheme

• US 9,454,433 B2
• Filed: 06/11/2014
• Issued: 09/27/2016
• Est. Priority Date: 04/17/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Dennett, Christopher M., Toussi, Farnaz
• Primary Examiner(s): Wilson, Yolanda L
• Application Number:
  US14/301,641
• Publication Number:
  US 20150301895A1
• Time in Patent Office: 839 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk

Abstract

Method and apparatus for redundant array of independent disks (RAID) recovery are disclosed. In one embodiment, a RAID controller schedules requests to rebuild failed drives based on the wear state of secondary drives and input/output (I/O) activity. The controller may be configured to assign higher scheduling priority to rebuild requests only when necessary, so as to reduce the time needed for the rebuild and to avoid affecting performance of the RAID system. In particular, the controller may give higher priority to rebuild requests if secondary drive failure is likely. In addition, the controller may determine when write-intensive periods occur, and assign lower priority to rebuild requests during such periods.

Claims

1. A method for data recovery in a storage system in which fault tolerance in the storage system is provided using multiple copies of data, comprising:

receiving a request to rebuild a failed drive in the storage system;

determining states of one or more secondary drives associated with the rebuild request;

if one or more of the secondary drives are in a failing state, prioritizing a rebuild operation over read and write operations; and

if none of the secondary drives are in a failing state, delaying the rebuild operation if the storage system is in a write-intensive period.

2. The method of claim 1, further comprising,

maintaining a counter for a number of dispatched write requests, a counter for a number of dispatched read requests, and a rebuild age for the rebuild request;

and

maintaining a queue of read requests, a queue of write requests, and a queue of rebuild requests.

3. The method of claim 2, further comprising, upon determining that only rebuild requests are outstanding, dispatching the outstanding rebuild requests and resetting the counter for the number of dispatched write requests and the counter for the number of dispatched read requests if the counter values are greater than 0.

4. The method of claim 2, wherein the rebuild operation is delayed, and the rebuild age incremented, until the rebuild age exceeds a threshold age.

5. The method of claim 2, wherein the write-intensive period is determined based on whether the number of write requests indicated by the counter for the number of dispatched write requests is greater than the number of read requests indicated by the counter for the number of dispatched read requests.

6. The method of claim 1, wherein delaying the rebuild operation includes assigning a scheduling priority to the rebuild request that is lower than scheduling priorities assigned to read and write requests.

7. The method of claim 1, wherein the storage system includes one or more solid-state drives (SSDs).

8. The method claim 1, wherein the states of the secondary drives are determined based on wear status reports issued by drives of the storage system.