FAILOVER METHOD, PROGRAM, FAILOVER APPARATUS AND FAILOVER SYSTEM

• US 20090271655A1
• Filed: 04/20/2009
• Published: 10/29/2009
• Est. Priority Date: 04/23/2008
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: MASUDA, Kouji, YOSHIMURA, Satoshi, HOTTA, Shinji, KUNIMI, Toshiki
• Granted Patent: US 8,010,830 B2
• Time in Patent Office: Days
• US Class Current: 714/4
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence

Abstract

The invention is to provide a failover method for improving overall reliability of a system including a large number of servers. Priorities set based on operating statuses of respective servers and shared devices capable of communicating with the respective servers in a failover system are stored in a storage portion while the priorities are associated with the respective servers, so that a failure processing portion of a management server retrieves a server based on the priorities and designates the retrieved server as a spare server when a failure occurs in one of the servers.

Claims

1. A failover method in a failover apparatus provided in a system with servers capable of communicating with one another, for taking over processing to a spare server selected from other servers than a server executing the processing when a failure occurs in the server executing the processing, wherein:

server priorities set based on operating statuses of the respective servers and devices capable of communicating with the respective servers in the system are stored in a storage portion of the failover apparatus so that the priorities are associated with the respective servers; and

a processing portion of the failover apparatus retrieves a server based on the priorities stored in the storage portion and designates the retrieved server as the spare server when the failure occurs.

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

retrieval policies describing preference items for server retrieval are stored in the storage portion;

the priorities are updated based on the preference items determined by the retrieval policies; and

the server retrieval is performed based on the updated priorities.

3. A failover method according to claim 2, wherein the retrieval policies are at least one of preference to a higher probability of operation recovery, preference to a server smaller in the number of devices connected to the server, preference to a server higher in redundancy of devices connected to the server, preference to a server placed in the same chassis, and preference to a server placed in another chassis.

4. A failover method according to claim 2, wherein the retrieval policies are designed so that a plurality of retrieval policies can be set for each server.

5. A failover method according to claim 1, wherein the failover apparatus acquires information about operating statuses from devices of the system at a predetermining timing and updates the priorities based on the acquired information.

6. A failover method according to claim 5, wherein the predetermined timing is when the failure occurs.

7. A failover method according to claim 5, wherein the predetermined timing is when any one of devices included in the system is changed or when the operating status of any one of devices included in the system is changed.

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

the servers are logical servers each provided as a set of servers; and

the priorities are set for the logical servers respectively.

9. A failover method according to claim 8, wherein the priority set for each logical server is set based on priorities of servers which are constituent members of the logical server.

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

performance information about respective devices of the system is stored in the storage portion so that the performance information is associated with the respective devices; and

the spare server is retrieved based on the priorities and the performance information when the retrieval is required.

11. A failover method according to claim 10, wherein the performance information is at least one member selected from the group consisting of the number of other devices connected to each device, and the communication speed in each device.

12. A failover method according to claim 1, wherein the devices connected to the servers are at least one member selected from the group consisting of Fibre Channel switches, network switches, and chassis.

13. A program for making a computer execute a failover method according to claim 1.

14. A failover apparatus in a system with servers capable of communicating with one another, for taking over processing to a spare server selected from other servers than a server executing the processing when a failure occurs in the server executing the processing, the failover apparatus comprising:

a storage portion in which server priorities set based on operating statuses of the respective servers and devices capable of communicating with the servers in the system are stored while the priorities are associated with the respective servers;

a server retrieval portion which retrieves a server based on the priorities stored in the storage portion when the failure occurs; and

a failover processing portion which designates the retrieved server as the spare server.

15. A failover apparatus according to claim 14, wherein:

retrieval policies describing preference items for server retrieval are stored in the storage portion;

the priorities are updated based on the preference items determined by the retrieval policies; and

the server retrieval portion performs the server retrieval based on the updated priorities.

16. A failover apparatus according to claim 14, wherein:

the servers are logical servers each provided as a set of servers; and

the priorities are set for the logical servers respectively.

17. A failover apparatus according to claim 14, wherein:

performance information about devices in the system is stored in the storage portion so that the performance information is associated with the devices; and

the server retrieval portion retrieves the spare server based on the priorities and the performance information when the retrieval is required.

18. A failover system comprising servers capable of communicating with one another, a failover apparatus which takes over processing to a spare server selected from other servers than a server executing the processing when a failure occurs in the server executing the processing, and devices capable of communicating with the respective servers, wherein:

the failover apparatus includes a storage portion in which server priorities set based on operating statuses of the servers and the devices capable of communicating with the respective servers in the system are stored so that the priorities are associated with the respective servers;

the failover apparatus retrieves a server based on the priorities stored in the storage portion when the failure occurs; and

the failover apparatus designates the retrieved server as the spare server.