Backup system and backup method

• US 20070234115A1
• Filed: 06/08/2006
• Published: 10/04/2007
• Est. Priority Date: 04/04/2006
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Saika, Nobuyuki
• Granted Patent: US 7,487,390 B2
• Time in Patent Office: Days
• US Class Current: 714/13
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/203 using migration

Abstract

Provided is a method causing a node of a local site or of a remote site to take over a file server when a failure occurs to a node of the local site. The method includes the steps of: setting a priority of the failover to the respective nodes of the local site; setting a failover destination of the respective nodes of the local site to a node of the remote site or the local site based on correspondences between the priority and an extent of the failover destination set in advance; backing up data to respective nodes of the determined failover destination; causing the failover destinations and nodes to mutually detect a failure; and causing a node of the failover destination to take over a task if the node of the failover destination detects a failure.

Claims

1. A failover method used for a system,

the system including:

a primary site having;

multiple nodes that provide a task; and

a first storage system allocated to the nodes; and

a secondary site having;

multiple nodes that can provide a task; and

a second storage system allocated to the nodes,

the method allowing, if a failure occurs to a node of the primary site, the task to be taken over by a node of the primary site or a node of the secondary site,

the method comprising the steps of;

setting a priority of a failover for the respective nodes of the primary site;

setting a failover destination of the respective nodes of the primary site to one of a node of the secondary site and a node of the primary site based on a correspondence set in advance between a priority and an extent of the failover destination;

backing up data from a node of the primary site to a node of a determined failover destination;

causing the node of the failover destination and the node of the primary site to mutually detect a failure; and

causing the node of the failover destination to take over a task if the node of the failover destination detects a failure of the node of the primary site.

2. A failover method according to claim 1, wherein:

the setting the priority sets a first priority or a second priority lower than the first priority based on availability of the node; and

the setting the failover destination of the respective nodes of the primary site to a node of the secondary site or a node of the primary site sets the secondary site to the extent of the failover destination when the first priority is set to the node, and sets the primary site to the extent of the failover destination when the second priority is set to the node.

3. A failover method according to claim 2, further comprising the steps of:

causing the secondary site to detect the number of failures that have occurred in the primary site; and

causing the secondary site to take over in advance a task of a node of the primary site to which the first priority is set even if a failure is not occurring to the node when the number of the failures exceeds a predetermined upper limit.

4. A failover method according to claim 3, wherein:

the causing the secondary site to take over a task in advance comprises the steps of;

causing the node of the secondary site to take over the task of the node of the primary site to which the first priority is set; and

causing a node of the secondary site to take over a task of a node of the primary site to which the second priority is set regardless of a presence of a failure after a predetermined grace period has elapsed.

5. A failover method according to claim 3, wherein:

the causing the secondary site to detect the number of failures that have occurred in the primary site comprises the steps of;

chronologically comparing the identifier of a task provided by a node with physical identifiers of the node; and

detecting the number of the failures based on the number of previous physical identifiers of the node in consistent with the physical identifier of the node corresponding to the identifier of the task.

6. A failover method according to claim 1, wherein:

the first storage system is an NAS system that provides the multiple nodes of the primary site with a file system, thereby assigning a storage volume to the respective nodes; and

the node is set on a logical partition obtained by logically partitioning one computer resource.

7. A failover method according to claim 2, wherein:

the setting a failover destination of the respective nodes of the primary site to one of a node of the secondary site and a node of the primary site comprises the steps of;

searching for the failover destination in an extent corresponding to the priority set to the node; and

extending the extent of the failover destination by changing the priority if there is no node to take over the task within the extent.

8. A failover method according to claim 1, wherein:

the primary site and the secondary site are connected with each other via a network;

at least one of the first storage system and the second storage system comprises a shared storage volume which can be accessed via the network;

the shared storage volume stores information on the respective nodes of the primary site whose task is to be taken over; and

the causing the node of the failover destination to take over a task obtains the information on the respective nodes of the primary site whose task is to be taken over, from the shared storage volume, and causing the node to take over the task based on the information on the node.

9. A failover method according to claim 1, wherein:

the system further comprises a tertiary site;

the tertiary site takes over a task of a node of the primary site when a failure occurs to the node of the primary site; and

the tertiary site takes over a task in place of the secondary site when a failure occurs to a node of the primary site.

10. A backup system, comprising:

a primary site comprising;

multiple nodes that provide a task; and

a first storage system allocated to the corresponding nodes;

a secondary site comprising;

multiple nodes that provide a task; and

a second storage system allocated to the corresponding nodes; and

a network connecting the primary site and the secondary site with each other;

wherein;

a node of the secondary site takes over a task of a node of the primary site if a failure occurs to the node of the primary site;

the node of the primary site comprises;

a priority setting unit for setting a priority of a failover for the respective nodes of the primary site;

a cluster setting unit for setting a failover destination of the respective nodes of the primary site to one of a node of the secondary site and a node of the primary site based on the specified priority;

a first failure detecting unit for cooperating with a node of the specified failover destination to monitor a failure;

a backup processing unit for baking up data of itself to the node of the specified failover destination; and

a first failover processing unit for taking over a task of the node to be monitored if the first failure detecting unit detects a failure; and

the node of the secondary site comprises;

a second failure detecting unit for cooperating with a node of a failover source to monitor a failure; and

a second failover processing unit for causing the specified node to take over a task of the monitored node if a failure is detected.

11. A backup system according to claim 10, wherein:

the priority setting unit sets a first priority or a second priority lower than the first priority based on availability of the node of the primary site; and

the cluster setting unit sets the secondary site to the extent of the failover destination when the first priority is set to the node, and sets the primary site to the extent of the failover destination when the second priority is set to the node.

12. A backup system according to claim 11, wherein:

the node of the secondary site comprises a backup source monitoring unit for detecting the number of failures that have occurred in the primary site; and

the second failover processing unit takes over in advance a task of a node of the primary site to which the first priority is set even if a failure is not occurring to the node when the number of failures exceeds a predetermined upper limit.

13. A backup system according to claim 12, wherein:

the second failover processing unit takes over a task of a node of the primary site to which the second priority is set regardless of a presence of a failure of the node of the primary site after a predetermined grace period has elapsed since that start of the takeover of a task of a node of the primary site to which the first priority is set.

14. A backup system according to claim 12, wherein:

the backup source monitoring unit chronologically compares an identifier of a task provided by a node of the primary site with physical identifiers of the node, and detects the number of the failures based on the number of previous physical identifiers of the node in consistent with the physical identifier of the node corresponding to the identifier of the task.

15. A backup system according to claim 10, wherein:

the first storage system is an NAS system that provides the multiple nodes of the primary site with a file system, thereby assigning a storage volume to the respective nodes; and

the node is set on a logical partition obtained by logically partitioning one computer resource.

16. A backup system according to claim 11, wherein:

the cluster setting unit searches for the failover destination in an extent corresponding to the priority set to the node of the primary site, and extends the extent of the failover destination by changing the priority if there is no node to take over the task within the extent.

17. A backup system according to claim 10, wherein:

at least one of the first storage system and the second storage system comprises a shared storage volume which can be accessed via the network;

the cluster setting unit stores information on the respective nodes of the primary site whose task is to be taken over, in the shared storage volume; and

the second failover processing unit obtains the information on the respective nodes of the primary site whose task is to be taken over, from the shared storage volume, and causing the node to take over the task based on the information on the node.