STORAGE APPARATUS

• US 20120151169A1
• Filed: 12/08/2010
• Published: 06/14/2012
• Est. Priority Date: 12/08/2010
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Mori, Hajime, Kawakami, Norihiko, Hirayama, Hiroshi, Ouchi, Katsumi
• Granted Patent: US 8,463,981 B2
• Time in Patent Office: Days
• US Class Current: 711/166
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0641 De-duplication techniques

Abstract

It is an object of the present invention to provide a storage apparatus that can suppress access performance degradation owing to processing of data deduplication. The storage apparatus according to the present invention acquires a sum of the access frequencies to each of redundantly allocated pieces of data, and performs the deduplication on the pieces of data having the sum of access frequencies less than a prescribed threshold (FIG. 10).

Claims

1. A storage apparatus, comprising:

a storage device storing data;

a controller controlling data reading and writing of the storage device;

a volume virtualization unit that virtualizes the storage device and provides a higher-level device with the virtualized storage device as a virtual storage capacity;

a administration table administrating which location in the storage device a piece of data in the virtual storage capacity is stored;

an access frequency administration table administrating an access frequency to the piece of data stored in the storage device; and

a deduplication unit that performs deduplication by integrating the pieces of data redundantly allocated in the storage device,

wherein the deduplication unit acquires the access frequencies to the respective pieces of data redundantly allocated in the storage device from the access frequency administration table, obtains a sum thereof, and, if the sum is less than a prescribed threshold,

performs the deduplication by integrating the pieces of data into any one thereof and

updates the storage location information of the piece of data in the storage device before the deduplication that is held by the administration table to storage location information of the piece of data in the storage device after the deduplication.

2. The storage apparatus according to claim 1,

wherein the deduplication unit,

if the access frequency to the piece of data equals to or more than the prescribed threshold after the pieces of data are integrated in the deduplication,

duplicates the piece of data subjected to the deduplication and the pieces of data are redundantly allocated again, and

updates the storage location information of the piece of data in the storage device after the deduplication that is held by the administration table to storage location information in the storage device after the duplication.

3. The storage apparatus according to claim 1,

wherein the deduplication unit

acquires an access performance sum of at least one of the storage devices provided as the virtual storage capacity by the volume virtualization unit, and

sets the prescribed threshold to a value less than or equal to the access performance sum.

4. The storage apparatus according to claim 2,

further comprising a deduplication administration table that holds identification information of the piece of data subjected to the deduplication, wherein

the deduplication unit

identifies the piece of data to be a target of the duplication according to the identification information held by the deduplication administration table.

5. The storage apparatus according to claim 4,

wherein the deduplication administration table holds the value of the access frequency held by the access frequency administration table in a manner divided with respect to each virtual storage capacity, and

the deduplication unit,

when duplicating the piece of data subjected to the deduplication and redundantly allocating the duplicated piece of data again,

regards, as the target of duplication, the piece of data corresponding to the highest access frequency among access frequencies held by the deduplication administration table in a manner divided with respect to each virtual storage capacity.

6. The storage apparatus according to claim 4,

wherein the deduplication administration table holds the access frequency to the piece of data stored in the storage device in a manner divided with respect to each virtual storage capacity, and

the deduplication unit,

when duplicating the piece of data subjected to the deduplication and redundantly allocating the duplicated piece of data again,

sets the value of the access frequency corresponding to the piece of data before the duplication that is held by the deduplication administration table, as an access frequency corresponding to the related piece of data of the access frequency administration table.

7. The storage apparatus according to claim 4,

wherein the deduplication administration table holds the access frequency to the piece of data stored in the storage device in a manner divided with respect to each virtual storage capacity, and

the deduplication unit,

when duplicating the piece of data subjected to the deduplication and redundantly allocating the duplicated piece of data again,

sets the value of the access frequency corresponding to the piece of data in the access frequency administration table to an unconfirmed state.

8. The storage apparatus according to claim 1,

wherein the access frequency administration table administrates a writing access frequency to the piece of data stored in the storage device as the access frequency.

9. The storage apparatus according to claim 1,

wherein the access frequency administration table administrates a sum of a writing access frequency and a reading access frequency to the piece of data stored in the storage device as the access frequency.

10. The storage apparatus according to claim 1,

further comprising a deduplication target data administration table that holds a list of the piece of data to be a target on which the deduplication is performed, wherein

the deduplication unit

preliminarily extracts the piece of data to be the target of the deduplication according to the access frequency administrated by the access frequency administration table before the deduplication is performed, and stores the list of the extracted piece of data in the deduplication target data administration table,

determines, only for the piece of data held by the deduplication target data administration table, whether the piece of data is redundantly allocated in multiple of the storage device or not, and,

if the piece of data is redundant, the deduplication is performed on that piece of data.

11. The storage apparatus according to claim 10,

wherein the deduplication unit

calculates a hash value of the piece of data held by the deduplication target data administration table and stores the hash value in the deduplication target data administration table, and,

if the hash values of the pieces of data allocated in the storage devices are identical with each other, determines that the pieces of data are redundantly allocated in the storage apparatuses.

12. The storage apparatus according to claim 11,

wherein the controller

determines whether or not writing to the piece of data listed on the deduplication target data administration table is performed before writing the piece of data in the storage device, and

the deduplication unit,

if the controller determines that the writing to the piece of data listed on the deduplication target data administration table is performed,

calculates the hash value again and stores the hash value in the deduplication target data administration table.

13. The storage apparatus according to claim 1,

wherein the volume virtualization unit

divides the storage devices into a high access frequency group having a high access frequency and a low access frequency group having a low access frequency, and

the deduplication unit

acquires the access frequency to the group from the access frequency administration table,

performs the deduplication on the pieces of data stored in the storage devices belonging to the low access frequency group, and

does not perform the deduplication on the pieces of data stored in the storage devices belonging to the high access frequency group.

14. The storage apparatus according to claim 1,

wherein the volume virtualization unit

divides the storage devices into a group omitted from deduplication targets that has a sufficient capacity for storing the pieces of data having the high access frequency that is not to be a target on which the deduplication unit performs the deduplication, and a deduplication target group for storing the pieces of data having the low access frequency,

the deduplication unit

acquires the access frequency to the piece of data from the access frequency administration table, and, if the value is less than the prescribed threshold, moves the piece of data to the storage device belonging to the deduplication target group, and, if the value is at least the prescribed threshold, moves the piece of data to the storage device belonging to the group omitted from deduplication targets,

performs the deduplication on the pieces of data stored in the storage devices belonging to the deduplication target group, and

does not perform the deduplication on the pieces of data stored in the storage devices belonging to the group omitted from deduplication targets.