Shifting wearout of storage disks

• US 10,001,931 B2
• Filed: 03/14/2016
• Issued: 06/19/2018
• Est. Priority Date: 10/22/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Franke, Jeffery M., O'Connor, James A.
• Primary Examiner(s): Rones, Charles
• Assistant Examiner(s): Wong, Nanci N
• Application Number:
  US15/069,393
• Publication Number:
  US 20170115903A1
• Time in Patent Office: 827 Days
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/3034 where the computing system component is a storage system, e.g. DASD based or network based digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/34 Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
• CPC Class Codes: G06F 3/0616 in relation to life time, e.g. increasing Mean Time Between Failures [MTBF]
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0658 Controller construction arrangements
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

Technical solutions are described to forestall data loss caused by wearout of storage disks in an array of storage disks in a storage system by monitoring a rate of writes for a first storage disk in the array and determining a mean time to failure of the first storage disk. A start time is determined based on the mean time to failure, a number of storage disks in the array, and a time to replace a storage disk in the array. At the start time, a notification is issued as an alert to replace the first storage disk to forestall data loss caused by wearout of a second storage disk in conjunction with a wearout of the first storage disk.

Claims

1. A method for forestalling data loss caused by wearout of storage disks in an array of storage disks in a storage system, the method comprising:

monitoring, by a controller, a rate of writes for each storage disk in the array;

determining, by the controller, a predicted mean time (T1-sigma) to failure of the each storage disk based on the rate of writes;

determining, by the controller, a start time (Tstart) based on the predicted mean time to failure, a number of storage disks in the array (N), and an estimated time taken to replace a storage disk in the array (Trepair) by computing Tstart=N×

[Trepair+Twait] +T1-sigma, the Twait being a predetermined delay;

generating, by the controller, a schedule to issue notifications for replacement of the each storage disk, the schedule comprises a plurality of the determined start times corresponding to each respective storage disk in the array, a duration between successive start times including the predetermined delay; and

issuing, by the controller, a plurality of notifications, each of which is issued at the respective start time from the schedule, each notification includes an alert to replace the corresponding storage disk, wherein replacement of a first storage disk at the respective start time forestalls the data loss caused by a wearout of a second storage disk of the array in conjunction with a wearout of the first storage disk.

2. The method of claim 1, wherein the estimated time taken to replace a storage disk in the array includes a time to detect wearout of a storage disk, procure a replacement storage disk, replace the storage disk with the replacement disk, and rebuild the data in the storage system according to a recovery scheme of the storage system.

3. The method of claim 1, wherein the start time is a first start time, the notification is a first notification, and the method further comprising:

after issuing the notification to replace the first storage disk;

determining, by the controller, a second start time based on the mean time to failure, the number of storage disks in the array, the time to replace a storage disk in the array, and a predetermined delay since replacement of the first storage disk; and

issuing, by the controller, at the second start time, a second notification, the second notification including an alert to replace the second storage disk, wherein replacing the second storage disk at the second start time forestalls the data loss caused by a wearout of a third storage disk of the array in conjunction with a wearout of the second storage disk.

4. The method of claim 1, wherein the start time to issue the notification is prior to a number of writes for the first storage disk being lesser than a predetermined maximum writes for the first storage disk, and the notification is distinct from a predictive failure alert.

5. The method of claim 4, wherein the notification includes alerts to replace all the storage disks in the array in conjunction.

6. The method of claim 5, wherein a storage disk in the array is one of a solid state drive, a flash memory, or a flash dual in-line memory module (DIMM), interfaced using any one of a peripheral component interconnect (PCI) card, a serial attached small computer system interface (SAS).

7. The method of claim 6, wherein the array of storage disks is maintained as a redundant array of independent disks (RAID).

8. The method of claim 1, wherein a storage disk in the array is one of a solid state drive, a flash memory, or a flash dual in-line memory module (DIMM), interfaced using a peripheral component interconnect (PCI) card or a serial attached small computer system interface (SAS).

9. The method of claim 1, wherein the array of storage disks is maintained as a redundant array of independent disks (RAID).