REDUCTION OF I/O LATENCY FOR WRITABLE COPY-ON-WRITE SNAPSHOT FUNCTION

• US 20110208909A1
• Filed: 02/24/2010
• Published: 08/25/2011
• Est. Priority Date: 02/24/2010
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: KAWAGUCHI, Tomohiro
• Granted Patent: US 8,447,943 B2
• Time in Patent Office: Days
• US Class Current: 711/113
• CPC Class Codes: G06F 11/1458 Management of the backup or restore process
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

According to one aspect of the invention, a method of controlling a storage system comprises storing data in a first volume in the storage system which has volumes including the first volume and a plurality of second volumes; prohibiting write I/O (input/output) access against the first volume after storing the data in the first volume; performing subsequent write requests received by the storage system against the second volumes in the storage system after storing the data in the first volume, each write request having a target volume which is one of the second volumes; and in response to each one write request of the write requests, determining whether the target volume of the one write request is write prohibited or not, and performing the one write request only if the target volume is not write prohibited.

Claims

1. A method of controlling a storage system, the method comprising:

storing data in a first volume in the storage system which has volumes including the first volume and a plurality of second volumes;

prohibiting write I/O (input/output) access against the first volume after storing the data in the first volume;

performing subsequent write requests received by the storage system against the second volumes in the storage system after storing the data in the first volume, each write request having a target volume which is one of the second volumes; and

in response to each one write request of the write requests, determining whether the target volume of the one write request is write prohibited or not, and performing the one write request only if the target volume is not write prohibited.

2. The method according to claim 1, further comprising:

setting a flag against a volume, indicating that write access to the volume is prohibited,

wherein the determining includes checking whether the flag is set against the target volume.

3. The method according to claim 1, further comprising:

receiving, by the storage system, a read request for one second volume of the plurality of second volumes; and

in response to the read request, providing an image of data obtained from the first volume and data transferred from the one second volume to a host computer.

4. The method according to claim 1,

wherein the second volumes are virtual volumes, which are allocated from a pool volume in units of pages in response to a write request, and

wherein writing to the pages in the second volumes for performing the subsequent write requests are performed asynchronously.

5. The method according to claim 4,

wherein data for the subsequent write requests are written to a cache memory, and

wherein writing to the pages from the cache memory are performed before a dirty area of the cache memory reaches a predetermined threshold.

6. The method according to claim 1,

wherein the subsequent write requests are for snapshot images.

7. The method according to claim 1,

wherein the subsequent write requests are sent from host computers;

and

wherein each host computer corresponds to one of the second volumes.

8. The method according to claim 1, further comprising, if the second volumes are virtual volumes which are allocated from a pool volume in units of pages in response to a write request, and if a page has not been allocated for the target volume of the write request:

running cache control to search cache area for related area of a paired logical volume which is paired with a virtual volume without locking the related area of the paired logical volume as exclusive status.

9. The method according to claim 8, further comprising:

if data of the related area is stored in a cache area of the paired logical volume, copying the data from the cache area of the paired logical volume to a cache area of the virtual volume;

if data of the related area is not stored in a cache area of the paired logical volume, copying the data from a disk of the paired logical volume to a cache area of the virtual volume; and

changing a virtual volume cache area status of the virtual volume to “

clean”

without a need to unlock the related area of the paired logical volume.

10. The method according to claim 1, further comprising, if the second volumes are virtual volumes which are allocated from a pool volume in units of pages in response to a write request, and if a page has not been allocated for the target volume of the write request:

allocating a snapshot pool page to a virtual volume page of a virtual volume;

searching cache area for the related area of a paired logical volume which is paired with a virtual volume;

checking whether data of the related area is stored in a cache area of the paired logical volume or not, and if yes, copying the data from the cache area of paired logical volume to a cache area of the allocated snapshot pool page, and if no, copying the data from a disk of logical volume area to a cache area of the allocated snapshot pool page; and

changing the cache status of the copy-on-write snapshot page to “

Dirty.”

11. The method according to claim 1,

wherein the data stored in the first volume is a Gold Image to be shared by all the host computers;

wherein each host computer is coupled with a corresponding second volume; and

wherein updates to each host computer are stored in the corresponding second volume.

12. A storage system in an information system that includes a plurality of host computers coupled via a network to the storage system, the storage system comprising:

a processor;

a memory;

a storage which has volumes including a first volume and a plurality of second volumes;

a storage controller configured to

store data in the first volume in the storage system;

prohibit write I/O access against the first volume after storing the data in the first volume;

perform subsequent write requests received by the storage system against the second volumes in the storage system after storing the data in the first volume, each write request having a target volume which is one of the second volumes; and

in response to each one write request of the write requests, determine whether the target volume of the one write request is write prohibited or not, and perform the one write request only if the target volume is not write prohibited.

13. The storage system according to claim 12,

wherein the storage controller is configured to set a flag against a volume, indicating that write access to the volume is prohibited,

wherein determining whether the target volume of the one write request is write prohibited or not includes checking whether the flag is set against the target volume.

14. The storage system according to claim 12, wherein the storage controller is configured to:

receive a read request for one second volume of the plurality of second volumes; and

in response to the read request, provide an image of data obtained from the first volume and data transferred from the one second volume to a host computer.

15. The storage system according to claim 12,

wherein the second volumes are virtual volumes, which are allocated from a pool volume in units of pages in response to a write request, and

wherein the storage controller is configured to write to the pages in the second volumes for performing the subsequent write requests asynchronously.

16. The storage system according to claim 15, wherein the storage controller is configured to:

write data for the subsequent write requests to a cache memory, and

write to the pages from the cache memory before a dirty area of the cache memory reaches a predetermined threshold.

17. The storage system according to claim 12,

wherein the subsequent write requests are for snapshot images.

18. The storage system according to claim 12,

wherein the subsequent write requests are sent from the host computers;

and

wherein each host computer corresponds to one of the second volumes.

19. A computer-readable storage medium storing a plurality of instructions for controlling a data processor to process I/O (input/output) in an information system that includes a plurality of host computers coupled via a network to the storage system, the plurality of instructions comprising:

instructions that cause the data processor to store data in a first volume in the storage system which has volumes including the first volume and a plurality of second volumes;

instructions that cause the data processor to prohibit write I/O access against the first volume after storing the data in the first volume;

instructions that cause the data processor to perform subsequent write requests received by the storage system against the second volumes in the storage system after storing the data in the first volume, each write request having a target volume which is one of the second volumes; and

instructions that cause the data processor to, in response to each one write request of the write requests, determine whether the target volume of the one write request is write prohibited or not, and performing the one write request only if the target volume is not write prohibited.

20. The computer-readable storage medium according to claim 19, wherein the plurality of instructions further comprise:

instructions that cause the data processor to set a flag against a volume, indicating that write access to the volume is prohibited,

wherein determining whether the target volume of the one write request is write prohibited or not includes checking whether the flag is set against the target volume.