COMPUTER PROGRAM, METHOD, AND APPARATUS FOR CONTROLLING DATA ALLOCATION

• US 20110191628A1
• Filed: 12/13/2010
• Published: 08/04/2011
• Est. Priority Date: 12/18/2009
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Sakurai, Hideki, NOGUCHI, Yasuo
• Granted Patent: US 8,433,947 B2
• Time in Patent Office: Days
• US Class Current: 714/6.2
• CPC Class Codes: G06F 11/16 Error detection or correction of the data by redundancy in hardware
• 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 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2082 Data synchronisation
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence

Abstract

A computer executes a data allocation control program to control allocation of data in a plurality of disk nodes. A redundancy restoration module executes a redundancy restoration procedure by commanding disk nodes to create a new copy of redundancy-lost data. An error message reception module receives a write error message indicating a write error in a storage device during the redundancy restoration procedure and records an identifier of the faulty storage device in an error record memory. A copy command module identifies accessed data in the faulty storage device and commands a relevant disk node to create a copy of the identified data in the faulty storage device in the case where there is no redundant copy of the data.

Claims

1. A computer-readable, non-transitory medium storing a data allocation control program for controlling data allocation in a plurality of disk nodes that manage data in storage devices connected thereto, the data allocation control program, when executed by a computer, causing the computer to perform a procedure, the procedure comprising:

executing a redundancy restoration procedure by commanding a first disk node that manages redundancy-lost data to create a new copy of the redundancy-lost data, the redundancy-lost data being data stored in a first storage device and lacking a copy thereof;

recording, upon receipt of a write error message indicating a write error encountered by a second storage device during the redundancy restoration procedure, an identifier of the second storage device in an error record memory; and

identifying the second storage device by consulting the error record memory, selecting data stored in the second storage device as copy target data, and commanding a second disk node managing the second storage device to create a copy of the copy target data in a third storage device having no write error when the third storage device is found to have no such copy of the copy target data.

2. The computer-readable, non-transitory medium according to claim 1, wherein the redundancy restoration procedure comprises selecting each piece of redundancy-lost data one by one, and commanding the first disk node managing the selected piece of redundancy-lost data to create a new copy thereof in a fourth storage device that is different from the first storage device containing the selected piece of redundancy-lost data and has no write error.

3. The computer-readable, non-transitory medium according to claim 1, wherein:

the redundancy restoration procedure comprises investigating the plurality of storage devices to find the redundancy-lost data therein, storing management data in a management data memory to indicate storage locations of the found redundancy-lost data, sequentially selecting each piece of redundancy-lost data indicated by the management data in the management data memory, and commanding the first disk node managing the selected piece of redundancy-lost data to create a new copy thereof; and

the commanding the second disk node comprises storing additional management data in the management data memory, upon completion of copying the copy target data, the additional management data describing the resulting copy of the copy target data as being redundancy-lost data.

4. The computer-readable, non-transitory medium according to claim 3, wherein the redundancy restoration procedure further comprises:

searching, upon completion of copying every piece of redundancy-lost data indicated by the management data in the management data memory, the error record memory to determine whether any storage device has encountered a write error;

adding management data in the management data memory for redundancy-lost data that is to be produced as a result of disabling access to the second storage device; and

continuing to restore data redundancy based on the added management data.

5. The computer-readable, non-transitory medium according to claim 1, wherein the procedure further comprises:

selecting, upon receipt of a data access error message from a device which indicates that the device has encountered a write error in an attempt to access data in the second storage device, the accessed data in the second storage device as copy target data;

producing, when a copy of the selected copy target data is created as a result of the commanding the second disk node, management data to indicate the copy of the selected copy target data as a new destination of access; and

sending the management data back to the device from which the data access error message has been received.

6. The computer-readable, non-transitory medium according to claim 1, wherein the procedure further comprises initiating the redundancy restoration procedure for redundancy-lost data that is to be produced as a result of disabling access to a fourth storage device, in response to a write error message indicating a write error in the fourth storage device during a period when there is no ongoing redundancy restoration procedure.

7. A method executed by a computer to control data allocation in a plurality of disk nodes that manage data in storage devices connected thereto, the method comprising:

executing a redundancy restoration procedure by commanding a first disk node that manages redundancy-lost data to create a new copy of the redundancy-lost data, the redundancy-lost data being data stored in a first storage device and lacking a copy thereof;

recording, upon receipt of a write error message indicating a write error encountered by a second storage device during the redundancy restoration procedure, an identifier of the second storage device in an error record memory; and

identifying the second storage device by consulting the error record memory, selecting data stored in the second storage device as copy target data, and commanding a second disk node managing the second storage device to create a copy of the copy target data in a third storage device having no write error when the third storage device is found to have no such copy of the copy target data.

8. An apparatus which controls allocation of data in a plurality of disk nodes that manage data in storage devices connected thereto, the apparatus comprising:

a redundancy restoration module which executes a redundancy restoration procedure by commanding a first disk node that manages redundancy-lost data to create a new copy of the redundancy-lost data, the redundancy-lost data being data stored in a first storage device and lacking a copy thereof;

an error message reception module which records, upon receipt of a write error message indicating a write error encountered by a second storage device during the redundancy restoration procedure, an identifier of the second storage device in an error record memory; and

a copy command module which identifies the second storage device by consulting the error record memory, selects data stored in the second storage device as copy target data, and commands a second disk node managing the second storage device to create a copy of the copy target data in a third storage device having no write error when the third storage device is found to have no such copy of the copy target data.

9. An apparatus which controls allocation of data in a plurality of disk nodes that manage data in storage devices connected thereto, the apparatus comprising:

a processor configured to execute a procedure, the procedure comprising;

executing a redundancy restoration procedure by commanding a first disk node that manages redundancy-lost data to create a new copy of the redundancy-lost data, the redundancy-lost data being data stored in a first storage device and lacking a copy thereof;

recording, upon receipt of a write error message indicating a write error encountered by a second storage device during the redundancy restoration procedure, an identifier of the second storage device in an error record memory; and

identifying the second storage device by consulting the error record memory, selects data stored in the second storage device as copy target data, and commands a second disk node managing the second storage device to create a copy of the copy target data in a third storage device having no write error when the third storage device is found to have no such copy of the copy target data.