SERVER IMAGE MIGRATION

• US 20110071983A1
• Filed: 09/23/2009
• Published: 03/24/2011
• Est. Priority Date: 09/23/2009
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Murase, Atsushi
• Granted Patent: US 8,498,997 B2
• Time in Patent Office: Days
• US Class Current: 707/649
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5088 involving task migration

Abstract

A system includes a first site having a first server and a first storage system, a second site including a second server and a second storage system, and a third site coupled to the first and second sites, the third site having gold images. A method of migrating a virtual server image comprises storing a first gold image at the first site; storing a second gold image at the second site; creating a first snapshot volume at the first site based on the first gold image; sending the first gold image ID, update data of the first gold image, and address information for the update data, from the first site to the second site; creating a second snapshot volume at the second site based on the second gold image corresponding to the first gold image ID; allocating storage for storing the update data at the second site with reference to the address information; and storing the update data in the allocated storage.

Claims

1. A method of migrating a virtual server image in a system which includes a first site, a second site, and a third site, the first site having a first server and a first storage system coupled to the server via a first network, the second site including a second server and a second storage system coupled to the second server via a second network, the third site being coupled to the first site and the second site, the third site having a plurality of gold images, the virtual server image being migrated using the gold images, the method comprising:

storing a first gold image at the first site;

storing a second gold image at the second site;

creating a first snapshot volume at the first site based on the first gold image, the first gold image corresponding to a first gold image ID;

sending the first gold image ID, update data of the first gold image, and address information for the update data, from the first site to the second site;

creating a second snapshot volume at the second site based on the second gold image corresponding to the first gold image ID;

allocating storage for storing the update data at the second site with reference to the address information; and

storing the update data in the allocated storage.

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

providing a snapshot volume table which maintains at least information of a snapshot volume ID of the first snapshot volume and the first gold image ID of the first gold image.

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

upon receiving a request to write data to the first snapshot volume, allocating a chunk to a specified address of the first snapshot volume.

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

providing a chunk table which maintains information of a chunk ID of the allocated chunk, a snapshot volume ID of the first snapshot volume, a start LBA (logical block address) of the specified address, and a number of blocks of the allocated chunk.

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

allocating a virtual WWN (world wide name) to the first snapshot volume at the first site;

sending the virtual WWN to the second site; and

assigning the virtual WWN to the second snapshot volume at the second site.

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

checking whether a gold image corresponding to the first gold image ID is at the second site; and

if the gold image corresponding to the first gold image ID is not at the second site, obtaining a third gold image corresponding to the first gold image ID from the third site, which is to be stored as the second gold image at the second site, by specifying the first gold image ID.

7. The method according to claim 1,

wherein the system includes the first site to an N-th site each having a server and a storage system coupled to the server via a network;

wherein the first site has a plurality of gold images each corresponding to a gold image ID; and

wherein each of the sites except the first site and the third site periodically obtains gold images corresponding to the gold image IDs from the third site by specifying the gold image IDs corresponding to the gold images at the first site.

8. The method according to claim 1, wherein storing the first gold image at the first site comprises:

checking whether a portable media is designated as a source for the first gold image;

if the portable media is the source, obtaining the first gold image from the portable media; and

if the portable media is not the source, obtaining the first gold image from the third site.

9. The method according to claim 1, wherein the first gold image is a first generation first gold image, the first snapshot volume is a first generation first snapshot volume, and the update data is first generation update data, the method further comprising:

at an i-th generation starting with i=2 until i=N, after the i−

1-th update data is written to the i−

1-th generation first snapshot volume, using the i−

1-th generation first snapshot volume as an i-th generation gold image to create an i-th generation first snapshot volume at the first site, so as to create an N-th generation first snapshot volume at the first site based on an N-th generation gold image; and

sending the first gold image ID, the first generation to N-th generation update data, and address information for the first generation to N-th generation update data, from the first site to the second site.

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

providing a snapshot volume table which maintains at least information of an i-th generation first snapshot volume ID of the i-th generation first snapshot volume and an i-th generation first gold image ID of the i-th generation first gold image associated with the i-th generation first snapshot volume, from i=1 to i=N.

11. The method according to claim 1, wherein the first gold image is a first generation first gold image, the first snapshot volume is a first generation first snapshot volume, and the update data is first generation update data, the method further comprising:

at an i-th generation starting with i=2 until i=N, after the i−

1-th update data is written to the i−

1-th generation first snapshot volume, using the i−

1-th generation first snapshot volume as an i-th generation gold image to create an i-th generation first snapshot volume at the first site, so as to create an N-th generation first snapshot volume at the first site based on an N-th generation gold image;

combining all i-th generation update data to form a cumulative update data;

storing the cumulative update data in the first storage system; and

sending the first gold image ID, the cumulative update data, and address information for the cumulative update data, from the first site to the second site.

12. A method of migrating a virtual server image in a system which includes a first site, a second site, and a third site, the first site having a first server and a first storage system coupled to the server via a first network, the second site including a second server and a second storage system coupled to the second server via a second network, the third site being coupled to the first site and the second site, the third site having a plurality of gold images, the virtual server image being migrated using the gold images, the method comprising:

checking whether there is update data for the virtual server image to be migrated;

if there is update data, then

storing a first gold image at the first site;

creating a first snapshot volume at the first site based on the first gold image, the first gold image corresponding to a first gold image ID;

sending the first gold image ID, the update data of the first gold image, and address information for the update data, from the first site to the second site;

storing a second gold image at the second site, the second gold image corresponding to the first gold image ID;

creating a second snapshot volume at the second site based on the second gold image;

allocating storage for storing the update data at the second site with reference to the address information; and

storing the update data in the allocated storage; and

if there is no update data, then

storing a first gold image at the first site;

creating a first snapshot volume at the first site based on the first gold image, the first gold image corresponding to a first gold image ID;

sending the first gold image ID, from the first site to the second site;

storing a second gold image at the second site, the second gold image corresponding to the first gold image ID; and

creating a second snapshot volume at the second site based on the second gold image.

13. The method according to claim 12, further comprising:

providing a snapshot volume table which maintains at least information of a snapshot volume ID of the first snapshot volume and the first gold image ID of the first gold image.

14. The method according to claim 12, further comprising:

checking whether a gold image corresponding to the first gold image ID is at the second site; and

if the gold image corresponding to the first gold image ID is not at the second site, obtaining a third gold image corresponding to the first gold image ID from the third site, which is to be stored as the second gold image at the second site, by specifying the first gold image ID.

15. A system for migrating a virtual server image, the system comprising:

a first site having a first server and a first storage system coupled to the server via a first network;

a second site including a second server and a second storage system coupled to the second server via a second network; and

a third site being coupled to the first site and the second site, the third site having a plurality of gold images, the virtual server image being migrated using the gold images;

wherein the first storage system stores a first gold image, and creates a first snapshot volume based on the first gold image, the first gold image corresponding to a first gold image ID;

wherein the first site sends the first gold image ID, update data of the first gold image, and address information for the update data, to the second site;

wherein the second storage system stores a second gold image corresponding to the first gold image ID, creates a second snapshot volume based on the second gold image and the update data received at the second site, allocates storage for storing the update data at the second site with reference to the address information, and stores the update data in the allocated storage.

16. The system according to claim 15,

wherein the first storage system includes a snapshot volume table which maintains at least information of a snapshot volume ID of the first snapshot volume and the first gold image ID of the first gold image.

17. The system according to claim 15, wherein the first site:

checks whether a portable media is designated as a source for the first gold image;

if the portable media is the source, obtains the first gold image from the portable media; and

if the portable media is not the source, obtains the first gold image from the third site.

18. The system according to claim 15,

wherein the first gold image is a first generation first gold image, the first snapshot volume is a first generation first snapshot volume, and the update data is first generation update data;

wherein at an i-th generation starting with i=2 until i=N, after the i−

1-th update data is written to the i−

1-th generation first snapshot volume, the first storage system uses the i−

1-th generation first snapshot volume as an i-th generation gold image to create an i-th generation first snapshot volume at the first site, so as to create an N-th generation first snapshot volume at the first site based on an N-th generation gold image; and

wherein the first site sends the first gold image ID, the first generation to N-th generation update data, and address information for the first generation to N-th generation update data, from the first site to the second site.

19. The system according to claim 18,

wherein the first storage system provides a snapshot volume table which maintains at least information of an i-th generation first snapshot volume ID of the i-th generation first snapshot volume and an i-th generation first gold image ID of the i-th generation first gold image, from i=1 to i=N.

20. The system according to claim 15,

wherein the first gold image is a first generation first gold image, the first snapshot volume is a first generation first snapshot volume, and the update data is first generation update data;

wherein at an i-th generation starting with i=2 until i=N, after the i−

1-th update data is written to the i−

1-th generation first snapshot volume, the first storage system uses the i−

1-th generation first snapshot volume as an i-th generation gold image to create an i-th generation first snapshot volume at the first site, so as to create an N-th generation first snapshot volume at the first site based on an N-th generation gold image;

wherein the first storage system combines all i-th generation update data to form a cumulative update data, and stores the cumulative update data; and

wherein the first site sends the first gold image ID, the cumulative update data, and address information for the cumulative update data, to the second site.