DISK ARRAY DEVICE, OPERATING METHOD THEREOF AND PROGRAM-STORING MEDIUM

• US 20080244309A1
• Filed: 03/28/2008
• Published: 10/02/2008
• Est. Priority Date: 03/29/2007
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Fukuyama, Osanori
• Granted Patent: US 7,890,791 B2
• Time in Patent Office: Days
• US Class Current: 714/6
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/1662 the resynchronized component or unit being a persistent storage device re-synchronization of failed mirror storage G06F11/2082; rebuild or reconstruction of parity RAID storage G06F11/1008
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence

Abstract

Even if failure probabilities are different for hard disks due to the individual specificity such as a hard disk manufacturer, model number or the like, a disk with high failure probability is reliably determined and removed from the operating RAID to be kept in a standby state as a hot spare disk, thereby keeping a low failure probability of the disk array device. In order to realize this, the disk array device includes a disk controlling unit for, based on S.M.A.R.T. information of each of the hard disks read by a S.M.A.R.T. information reading unit, assigning a predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

Claims

1. A disk array device characterized by including a plurality of hard disks composing a RAID and a predetermined number of hard disks each arranged as a hot spare disk, the disk array device being characterized by comprising:

a S.M.A.R.T. information reading part for reading S.M.A.R.T. information of each of the hard disks included in the disk array device; and

a disk controller for, based on the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading part, assigning a predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

2. The disk array device of claim 1, wherein

a plurality of RAIDs each composed of the plurality of hard disks is arranged,

the disk array device further comprises an access frequency obtaining part for obtaining access frequency information to each of the RAIDs, and

the disk controller is configured to, based on the access frequency information to each of the RAIDs obtained by the access frequency obtaining part and the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading part, assign hard disks with lower failure probabilities to the RAIDs in descending order of access frequency of the RAIDs and assign a predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

3. The disk array device of claim 1, wherein in reexamination of which of the hard disks to be assigned to the hot spare disk, the disk controller calculates an earliest predicted failure time of each of the hard disks based on the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading part, and assigns a hard disk with a shortest time left until the predicted failure time to the hot spare disk.

4. The disk array device of claim 1, wherein the S.M.A.R.T. information reading part reads the S.M.A.R.T. information of each of the hard disks at predetermined fixed intervals.

5. A disk array device characterized by including a plurality of hard disks composing a RAID and a predetermined number of hard disks each arranged as a hot spare disk, the disk array device being characterized by comprising:

S.M.A.R.T. information reading means for reading S.M.A.R.T. information of each of the hard disks included in the disk array device; and

disk controlling means for, based on the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading means, assigning a predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

6. An operating method of a disk array device including a plurality of hard disks composing a RAID and a predetermined number of hard disks each arranged as a hot spare disk, the operating method being characterized by comprising:

a S.M.A.R.T. information reading step of reading S.M.A.R.T. information of each of the hard disks included in the disk array device; and

a disk controlling step of, based on the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading step, assigning the predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

7. The operating method of the disk array device of claim 6, wherein

a plurality of RAIDs each composed of the plurality of hard disks are arranged in the disk array device,

the operating method further comprises an access frequency obtaining step of obtaining access frequency information to each of the RAIDs, and

in the disk controlling step, the access frequency information to each of the RAIDs obtained by the access frequency obtaining step and the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading step are used as a basis to assign hard disks with lower failure probabilities to the RAIDs in descending order of access frequency of the RAIDs and to assign a predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

8. The operating method of the disk array device of claim 6, wherein in the disk controlling step, in reexamination of which of the hard disks to be assigned to the hot spare disk, the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading step is used as a basis to calculate an earliest predicted failure time of each of the hard disks and assign a hard disk with a shortest time left until the predicted failure time to the hot spare disk.

9. The operating method of the disk array device of claim 6, wherein in the S.M.A.R.T. information reading step, reading of the S.M.A.R.T. information of each of the hard disks is performed at predetermined fixed intervals.

10. A program-storing medium for storing an operating program of a disk array device including a plurality of hard disks composing a RAID and a predetermined number of hard disks each arranged as a hot spare disk, the program-storing medium being characterized by enabling a computer to execute:

S.M.A.R.T. information reading processing for reading S.M.A.R.T. information of each of the hard disks included in the disk array device; and

disk controlling processing for, based on the S.M.A.R.T. information of each of the hard disks read in the S.M.A.R.T. information reading processing, assigning a predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

11. The program-storing medium for storing the operating program of the disk array device of claim 10, wherein

a plurality of RAIDs each composed of the plurality of hard disks are arranged in the disk array device,

the program-storing medium further enables the computer to execute access frequency obtaining processing for obtaining access frequency information to each of the RAIDs, and

in the disk controlling processing, the access frequency information to each of the RAIDs obtained by the access frequency obtaining processing and the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading processing are used as a basis to assign hard disks with lower failure probabilities to the RAIDs in descending order of access frequency of the RAIDs and to assign the predetermined number of hard disks to the hot spare disk in descending order of failure probability of the hard disks.

12. The program-storing medium for storing the operating program of the disk array device of claim 10, wherein in the disk controlling processing, in reexamination of which of the hard disks to be assigned to the hot spare disk, the S.M.A.R.T. information of each of the hard disks read by the S.M.A.R.T. information reading processing is used as a basis to calculate an earliest predicted failure time of each of the hard disks and to assign a hard disk of shortest time left until the predicted failure time to the hot spare disk.

13. The program-storing medium for storing the operating program of the disk array device of claim 10, wherein in the S.M.A.R.T. information reading processing, reading of the S.M.A.R.T. information of each of the hard disks is performed at predetermined fixed intervals.