STORAGE SYSTEM AND DATA MANAGEMENT METHOD

• US 20100049754A1
• Filed: 10/01/2008
• Published: 02/25/2010
• Est. Priority Date: 08/21/2008
• Status: Abandoned Application

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: TAKAOKA, Nobumitsu, SUTOH, Atsushi
• Application Number:
  US12/243,004
• Publication Number:
  US 20100049754A1
• Time in Patent Office: Days
• US Class Current: N/A
• CPC Class Codes: G06F 11/1435 using file system or storage system metadata
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

In a NAS apparatus, a processor reads in, from a data volume, metadata of all files included in a file system at a base point-in-time of a snapshot of the data volume, and writes all the read-in metadata to an area of a difference data storage volume (difference volume), and in a storage apparatus, a difference data save processing unit, upon receiving a block write request from the latest base point-in-time to the subsequent base point-in-time, chronologically writes data stored in a block specified by the block write request to an area subsequent to the difference data storage volume area.

Claims

1. A storage system comprising:

a storage apparatus, which stores a volume that stores, for one or more files, a file system comprising real data and metadata that comprises the update time information of the files, and which receives a block write request that specifies a block of the volume; and

a file server, which receives from a computer a file write request that specifies a file, specifies a block of the volume in which the file specified by the file write request is stored, and sends a block write request that specifies the block of the specified volume to the storage apparatus,

wherein the file server has a write processing unit, which reads from the volume the metadata of all files included in the file system at a plurality of base points-in-time serving as bases for the restoration of the volume, and which sequentially writes all the read-in metadata to a prescribed difference data storage volume of the storage apparatus, and

the storage apparatus has a difference data save processing unit, which, upon receiving the block write request from the latest base point-in-time to the subsequent base point-in-time, chronologically writes the data stored in the block specified by the block write request to a storage area subsequent to the storage area in which the metadata of the difference data storage volume has been written.

2. The storage system according to claim 1, wherein the file server comprises:

an identification data receiving unit that receives identification data of a restore-targeted file;

a retrieval unit that retrieves metadata comprising the identification data by reading the storage area in which the metadata of the difference data storage volume is stored;

an acquisition unit that acquires the update time information of the restore-targeted file from the metadata when metadata comprising the identification data is capable of being retrieved by the retrieval unit; and

a presentation unit that presents a list related to the restore-targeted files comprising the acquired update time information.

3. The storage system according to claim 2, wherein the metadata comprises a plurality of inodes comprising block numbers of the volume in which real data corresponding to the respective files is stored, and the correspondence relationship between the identification data and the inodes, and

the file server further comprises:

a determination unit, which acquires a first inode that corresponds to the identification data of a first base point-in-time, and when the first inode exists without a second inode that corresponds to the identification data of the subsequent base point-in-time, determines that there is a possibility that the identification data of this inode has been changed, and

the presentation unit presents information showing that there is a possibility that the identification data has been changed, and update time information.

4. The storage system according to claim 2, wherein the file server further comprises:

a restore specification processing unit, which receives a specification from the list as to the update time of the restore-targeted file to be restored, and notifies the specification to the storage apparatus, and

the storage apparatus further comprises;

a restore processing unit that reads out data required to restore the restore-targeted file of the update time corresponding to the specification from the difference data storage volume, and restores the restore-targeted file.

5. The storage system according to claim 2, wherein the storage apparatus further comprises:

a semiconductor memory capable of storing data, the file server further comprises;

a cache controller that stores the metadata of all files at a plurality of the base points-in-time stored in the difference data storage volume in the semiconductor memory, and

the retrieval unit retrieves metadata comprising the identification data from the metadata stored in the semiconductor memory.

6. The storage system according to claim 4, further comprising:

an external device capable of storing data,

wherein the storage apparatus further comprises;

a save unit that saves data of the difference data storage volume to the external device, and

the restore processing unit reads out, from the external device, data required to restore the restore-targeted file in a state corresponding to the specification, and restores the restore-targeted file.

7. The storage system according to claim 6, wherein the storage apparatus further comprises:

a metadata maintenance unit that maintains the metadata in the difference data storage volume subsequent to saving the data of the difference data storage volume to the external device, and

the retrieval unit retrieves metadata comprising the identification data from the metadata of the difference data storage volume.

8. The storage system according to claim 1, wherein the difference data save processing unit of the storage apparatus uses the storage area of the difference data storage volume to create a virtual volume such that the metadata of all the files of one base point-in-time is stored in contiguous storage areas, and that data written by the difference data save processing unit is stored in a storage area subsequent to the metadata storage area from the one base point-in-time to the subsequent base point-in-time..

9. A data management method for a storage system that comprises a storage apparatus, which stores a logical volume that stores, for one or more files, a file system comprising real data and metadata that comprises the update time information of the files, and which receives a block write request that specifies a block of the logical volume;

and a file server, which receives from a computer a file write request that specifies a file, specifies a block of the logical volume in which the file specified by the file write request is stored, and sends a block write request that specifies the block of the specified logical volume to the storage apparatus, the data management method comprising;

write processing step of reading from the logical volume the metadata of all files included in the file system at a plurality of base points-in-time serving as bases for the restoration of the logical volume, and sequentially writing all the read-in metadata to a prescribed difference data storage volume of the storage apparatus; and

difference data save processing step of chronologically writing data stored in the block specified by the block write request to a storage area subsequent to the storage area to which the metadata of the difference data storage volume has been written when the storage apparatus receives the block write request from each base point-in-time to the subsequent base point-in-time.

10. The data management method according to claim 9, wherein the file server executes:

identification data receiving step of receiving identification data of a restore-targeted file;

retrieving step of retrieving metadata comprising the identification data by reading the storage area in which the metadata of the difference data storage volume is stored;

acquiring step of acquiring the update time information of the restore-targeted file from the metadata when metadata comprising the identification data is capable of being retrieved; and

presenting step of presenting a list related to the restore-targeted files comprising the acquired update time information.

11. The data management method according to claim 10, wherein the metadata comprises a plurality of inodes comprising a block number of the volume in which real data corresponding to respective files is stored, and the correspondence relationship between the identification data and the inodes, the data management method further comprising:

determining step of acquiring a first inode that corresponds to the identification data of a first base point-in-time, and when the first inode exists without a second inode that corresponds to the identification data of the subsequent base point-in-time, determining that there is a possibility that the identification data of this inode has been changed, and

the presenting step presents information showing that there is a possibility that the identification data has been changed, and update time information.

12. The data management method according to claim 10, further comprising:

restore specification processing step of receiving a specification from the list as to the update time of the restore-targeted file to be restored, and notifying the specification to the storage apparatus; and

restore processing step of reading out data required to restore the restore-targeted file of the update time corresponding to the specification from the difference data storage volume, and restoring the restore-targeted file.

13. The data management method according to claim 10, wherein the storage apparatus comprises a semiconductor memory capable of storing data, the data management method further comprising:

cache execution step of storing the metadata of all files at a plurality of the base points-in-time stored in the difference data storage volume in the semiconductor memory, and

the retrieving step retrieves metadata comprising the identification data from the metadata stored in the semiconductor memory.

14. The data management method according to claim 13, wherein the storage system further comprises an external device capable of storing data, the data management method further comprising:

saving step of saving data of the difference data storage volume to the external device, and

the restore processing step reads out, from the external device, data required to restore the restore-targeted file in a state corresponding to the specification, and restores the restore-targeted file.

15. The data management method according to claim 9, wherein the difference data save processing step uses a storage area of the difference data storage volume to create a virtual volume such that metadata of all files of one base point-in-time is stored in contiguous storage areas, and that the data of a block specified by the block write request is stored in a storage area subsequent to the metadata storage area from the one base point-in-time to the subsequent base point-in-time.