Dynamically adapting the fault tolerance and performance characteristics of a raid-based storage system by merging and splitting raid groups

• US 7,958,304 B1
• Filed: 04/30/2008
• Issued: 06/07/2011
• Est. Priority Date: 04/30/2008
• Status: Expired due to Fees

• Current Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Original Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Inventors: Gupta, Ratnesh, Leong, James, Goel, Atul
• Primary Examiner(s): BATAILLE, PIERRE MICHE
• Application Number:
  US12/113,065
• Time in Patent Office: 1,133 Days
• Field of Search: 711/112, 711/114, 711/170, 714/6, 714/E11.084, 714/E11.034
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1096 Parity calculation or recalculation after configuration or reconfiguration of the system

Abstract

Redundant Array of Inexpensive Disks (RAID) groups in a storage system are dynamically reconfigured by merging and splitting the RAID groups. When an indication of a change to data or system characteristics is received, disks in the RAID groups of the storage system can be reorganized to adapt to the change.

Claims

1. A computer-implemented method comprising:

receiving an indication of a change to at least one of data characteristics and system characteristics of a storage system, the change comprising a change in data criticality, wherein data criticality reflects an importance of data; and

in response to the change, dynamically reconfiguring Redundant Array of Inexpensive Disks (RAID) groups of the storage system by reorganizing storage devices in the RAID groups, wherein dynamically reconfiguring the RAID groups comprises;

if the indication is an increase in data criticality, dynamically merging a first RAID group with a second RAID group to form a merged RAID group that includes the storage devices of the first RAID group and of the second RAID group; and

if the indication is a decrease in data criticality, dynamically splitting one of the RAID groups into one or more split RAID groups, the one or more split RAID groups including the storage devices of the one of the RAID groups.

2. The method of claim 1, wherein dynamically reconfiguring RAID groups further comprises:

concatenating disk block numbers of the storage devices in at least two of the RAID groups to form an address space of a merged RAID group.

3. The method of claim 1, wherein dynamically reconfiguring RAID groups further comprises:

dynamically reconfiguring a first stripe of a first RAID group of the RAID groups concurrently with I/O operations on a second stripe of the first RAID group, wherein each of the first stripe and the second stripe contains data blocks from each of the storage devices in the first RAID group.

4. The method of claim 1, wherein dynamically reconfiguring RAID groups further comprises:

changing the number of the storage devices in a storage device container of the storage system in which the RAID groups reside, according to fault tolerance levels of the RAID groups before and after the reconfiguring.

5. The method of claim 1, wherein dynamically reconfiguring RAID groups further comprises:

changing a fault tolerance level of at least one of the RAID groups while maintaining the number of the storage devices in a storage device container of the storage system in which the RAID groups reside.

6. The method of claim 1, wherein dynamically reconfiguring RAID groups further comprises:

splitting an I/O operation on a first RAID group into multiple I/O operations during reconfiguration of the RAID groups, at least one of the multiple I/O operations performed on a stripe of the first RAID group that has completed the reconfiguration and at least another one of the multiple I/O operations performed on another stripe of the first RAID group that has not completed the reconfiguration.

7. A computer-implemented method comprising:

receiving an indication of a change in data criticality in a storage system, wherein data criticality reflects an importance of data;

reconfiguring Redundant Array of Inexpensive Disks (RAID) groups of the storage system by reorganizing storage devices in the RAID groups, wherein reconfiguring the RAID groups comprises;

if the indication is an increase in data criticality, dynamically merging a first RAID group with a second RAID group to form a merged RAID group that includes the storage devices of the first RAID group and of the second RAID group; and

if the indication is a decrease in data criticality, dynamically splitting one of the RAID groups into one or more split RAID groups, the one or more split RAID groups including the storage devices of the one of the RAID groups; and

performing I/O operations to the RAID groups concurrently with the reconfiguring.

8. The method of claim 7, wherein reconfiguring RAID groups further comprises:

changing the number of the storage devices in a storage device container of the storage system in which the RAID groups reside, according to fault tolerance levels of the RAID groups before and after the reconfiguring.

9. The method of claim 7, wherein reconfiguring RAID groups further comprises:

changing a fault tolerance level of at least one of the RAID groups while maintaining the number of the storage devices in a storage device container of the storage system in which the RAID groups reside.

10. The method of claim 7, wherein reconfiguring RAID groups further comprises:

creating a new RAID group while maintaining the number of the storage devices in the RAID groups and the fault tolerance level of the RAID groups.

11. A storage system comprising:

a plurality of storage devices, which include data storage devices to store data and parity storage devices to store parity bits of the data; and

a storage server coupled to the plurality of storage devices, the storage server to;

receive an indication of a change to at least one of data characteristics and system characteristics of a storage system, the change comprising a change in data criticality, wherein data criticality reflects an importance of data; and

in response to the change, dynamically reconfigure Redundant Array of Inexpensive Disks (RAID) groups of the storage system by reorganizing the data storage devices and the parity storage devices in the RAID groups, wherein dynamically reconfiguring the RAID groups comprises;

if the indication is an increase in data criticality, dynamically merging a first RAID group with a second RAID group to form a merged RAID group that includes the storage devices of the first RAID group and of the second RAID group; and

if the indication is a decrease in data criticality, dynamically splitting one of the RAID groups into one or more split RAID groups, the one or more split RAID groups including the storage devices of the one of the RAID groups.

12. The storage system of claim 11, wherein the storage server is to dynamically reconfigure a first stripe of a first RAID group of the RAID groups concurrently with I/O operations on a second stripe of the first RAID group, wherein each of the first stripe and the second stripe contains data blocks from each of the storage devices in the first RAID group.

13. The storage system of claim 11, wherein the server is to concatenate disk block numbers of the storage devices in at least two of the RAID groups to form an address space of a merged RAID group.

14. The storage system of claim 11, wherein the server is to change the number of the storage devices in a storage device container of the storage system in which the RAID groups reside, according to fault tolerance levels of the RAID groups before and after the reconfiguring.

15. The storage system of claim 11, wherein the server is to change a fault tolerance level of at least one of the RAID groups while maintaining the number of the storage devices in a storage device container of the storage system in which the RAID groups reside.