Cluster computing system and its failover method

• US 7,370,099 B2
• Filed: 08/28/2003
• Issued: 05/06/2008
• Est. Priority Date: 03/28/2003
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Hara, Norie, Achiwa, Kyosuke
• Primary Examiner(s): Caldwell; Andrew
• Assistant Examiner(s): Fritz; Bradford F.
• Application Number:
  US10/652,161
• Publication Number:
  US 20050005001A1
• Time in Patent Office: 1,713 Days
• Field of Search: 709/223, 709/224
• US Class Current: 709/223
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2048 where the redundant components share neither address space nor persistent storage
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A cluster computing system includes an active system and a standby system connected via wire networks. The active system and the standby system each has a wireless communications apparatus, such that when all of the wire networks connecting the active system with the standby system are disconnected, the wireless communications apparatus of the active system reports the standby system of the failure of the active system using its wireless communications capability with the wireless communications apparatus of the standby system. The report includes the verification ID of the active system, and the standby system begins operation upon receiving a packet containing the verification ID.

Claims

1. A cluster computing system comprising:

an active system having an active host computer and an active storage system, connected to said active host computer, for storing data accessible by said active host computer;

a standby system having a standby host computer and a standby storage system, connected to said standby host computer, for storing data accessible by said standby host computer;

a wireline network which connects said active and standby systems to each other,

wherein said wireline network includes a cluster network which connects the host computers of the active and standby systems to each other and an inter-storage network which connects the storage systems of the active and standby systems to each other;

an active wireless communications module, directly coupled to the active system, for performing wireless communications; and

a standby wireless communications module, directly coupled to the standby system, for performing wireless communications,

wherein said cluster network is used by each of said active and standby systems to monitor the status of the other of said active and standby systems,

wherein said each of said active and standby systems monitors, via said cluster network, the status of the other of said active and standby systems by transmitting status communications between said each of said active and standby systems and the other of said active and standby systems,

wherein said inter-storage network is used to conduct a remote copy function between said active and standby systems,

wherein said remote copy function operates, via said inter-storage network, such that when said active host computer writes data in said active storage system, a copy of said data is stored in said standby storage system,

wherein the active wireless communications module, directly coupled to the active system, monitors whether a failure has occurred in said active system and executes a first failure report to wirelessly report a failure of the active system to the standby wireless communications module, directly coupled to the standby system,

wherein the active wireless communications module, directly coupled to the active system, monitors whether a network disconnection state has occurred in the wireline network including said cluster network and said inter-storage network connecting the active system and the standby system and executes a second failure report to wirelessly report the network disconnection state has occurred in the wireline network to the standby wireless communications module, directly coupled to the standby system, and

wherein the second failure report is executed when both of the cluster network and the inter-storage network have the network disconnection state at the same time, and

wherein, upon receiving the first or second failure report from the active wireless communication module, directly coupled to the active system, the standby wireless communication module, directly coupled to the standby system, issues a predetermined operation processing instruction command to the standby host computer at the standby system based on whether the first failure report or the second failure report is received.

2. A cluster computing system according to claim 1, wherein the cluster network is an Internet Protocol (IP) network, and the inter-storage network is a Fibre Channel (FC) network.

3. A cluster computing system according to claim 1, wherein the active wireless communications module, directly coupled to the active system, wirelessly transmits a packet including a verification ID of the active system to the standby wireless communications module at the standby system, and upon receiving the packet including the verification ID, the standby wireless communications module, directly coupled to the standby system, recognizes the failure of the active system or the network disconnection state of the wireline network and performs operations based on the verification ID.

4. A cluster computing system according to claim 1, wherein, upon receiving a call instruction command from the active host computer at the active system, the active wireless communications module, directly coupled to the active system wirelessly communicates the first or second failure report to the standby wireless communications module, directly coupled to the standby system.

5. A cluster computing system according to claim 1, wherein each of the active wireless communication module, directly coupled to the active system and the standby wireless communication module, directly coupled to the standby system, is shielded from interference in its operation from a site failure of each of the active system and the standby system.

6. A cluster computing system comprising:

an active system having an active host computer and an active storage system, connected to said active host computer, for storing data accessible by said active host computer;

a standby system having a standby host computer and a standby storage system, connected to said standby host computer, for storing data accessible by said standby host computer;

a wireline network which connects said active and standby systems to each other,

wherein said wireline network includes a cluster network which connects the host computers of the active and standby systems to each other and an inter-storage network which connects the storage systems of the active and standby systems to each other;

an active wireless communications module, directly coupled to the active system, for performing wireless communications;

a standby wireless communications module, directly coupled to the standby system, for performing wireless communications;

a first module that keeps a system operation of the active system to continue when the active system cannot obtain information from the standby system via the wireline network; and

a second module that, upon detection of a failure of the active system, executes a first failure report to cause the active wireless communication module, directly coupled to the active system, to wirelessly report the failure of the active system to the standby wireless communications module, directly coupled to the standby system,

wherein said second module further, upon detection that a network disconnection state has occurred in said wireline network including said cluster network and said inter-storage network connecting the active system and the standby system, executes a second failure report to cause the active wireless communication module, directly coupled to the active system, to wirelessly report that the network disconnection state has occurred in the wireline network to the standby wireless communications module, directly coupled to the standby system,

wherein said cluster network is used by each of said active and standby systems to monitor the status of the other of said active and standby systems,

wherein said each of said active and standby systems monitors, via said cluster network, the status of the other of said active and standby systems by transmitting status communications between said each of said active and standby systems and the other of said active and standby systems,

wherein said inter-storage network is further used to conduct a remote copy function between said active and standby systems,

wherein said remote copy function operates, via said inter-storage network, such that when said active host computer writes data in said active storage system, a copy of said data is stored in said standby storage system,

wherein the active wireless communication module, directly coupled to the active system, wirelessly transmits the second failure report to the standby wireless communication module, directly coupled to the standby system, when said wireline network including both of the cluster network and the inter-storage network have the network disconnection state at the same time,

wherein the active wireless communication module, directly coupled to the active system, includes said second module, and

wherein, upon receiving the first failure report or the second failure report from the active wireless communication module, directly coupled to the active system, the standby wireless communication module, directly coupled to the standby system, issues a predetermined operation processing instruction command to the standby host computer at the standby system based on whether the first failure report or the second failure report is received.

7. A cluster computing system according to claim 6, further comprising:

a third module that, upon receiving the first failure report or the second failure report from the active wireless communication module, directly coupled to the active system, makes the standby wireless communication module, directly coupled to the standby system, start a predetermined operation processing based on whether the first failure report or the second failure report is received.

8. A cluster computing system according to claim 6, wherein the active wireless communications module, directly coupled to the active system, wirelessly transmits a packet including a verification ID of the active system to the standby wireless communications module, directly coupled to the standby system, and upon receiving the packet including the verification ID, the standby wireless communications module, directly coupled to the standby system, recognizes the failure of the active system or the network disconnection state of the wireline network and performs operations based on the verification ID.

9. A cluster computing system according to claim 6, wherein, upon receiving a call instruction command from the active host computer at the active operating system, the active wireless communications module, directly coupled to the active system, wirelessly communicates the first or second failure report to the standby wireless communications module, directly coupled to the standby system.

10. A cluster computing system according to claim 6, wherein the active wireless communication module, directly coupled to the active system, monitors a failure of the active host computer of the active system, and wirelessly communicates, upon detecting the failure, the first failure report to the standby wireless communications module, directly coupled to the standby system.

11. A cluster computing system according to claim 6, wherein each of the active wireless communication module, directly coupled to the active system, and the standby wireless communication module, directly coupled to the standby system, is shielded from influence of a site failure of each of the active system and the standby system.

12. A cluster computing system according to claim 7, wherein upon finding a failure on the active storage apparatus of the active system, the active host computer at the active system gives an instruction to the active wireless communication module, directly coupled to the active system, to wirelessly communicate the first failure report to the standby wireless communication module, directly coupled to the standby system.

13. A cluster computing system according to claim 12, wherein upon receiving the first failure report from the active wireless communication module, directly coupled to the active system, the standby wireless communication module, directly coupled to the standby system, instructs the standby host computer at the standby system to start said predetermined operation processing.

14. A failover method for a cluster computing system including an active system having an active host computer and an active storage system, connected to said active host computer, for storing data accessible by said active host computer, a standby system having a standby host computer and a standby storage system, connected to said standby host computer, for storing data accessible by said standby host computer, a wireline network which connects said active and standby systems to each other, an active wireless communications module, directly coupled to the active system and a standby wireless communications module, directly coupled to the standby system, each wireless communications module for performing wireless communications, the failover method comprising the steps of:

making the active system to continue a system operation even when the active system cannot obtain information from the standby system via the wireline network;

upon detection of a failure of the active storage system at the active system, making the active system to instruct the active wireless communication module, directly coupled to the active system, to wirelessly communicate a first failure report to report the failure to the standby wireless communications module directly coupled to the standby system;

upon receiving the first failure report from the active wireless communication module, directly coupled to the active system, making the standby wireless communication module, directly coupled to the standby system, to issue a first predetermined operation processing instruction command to the standby host computer at the standby system to start a first processing operation,

wherein said wireline network includes a cluster network which connects the host computers of the active and standby systems to each other and an inter-storage network which connects the storage systems of the active and standby systems to each other,

upon detection that a network disconnection state has occurred in said wireline network, making the active system to instruct the active wireless communication module, directly coupled to the active system, to wirelessly communicate a second failure report to report that the network disconnection state has occurred in said wireline network to the standby wireless communications module, directly coupled to the standby system; and

upon receiving the second failure report from the active wireless communication module, directly coupled to the active system, making the standby wireless communication module, directly coupled to the standby system, to issue a second predetermined operation processing instruction command to the standby host computer at the standby system to start a second processing operation,

wherein said first predetermined operation processing instruction command and said second predetermined operation processing instruction command are different from each other,

wherein said cluster network is used by each of said active and standby systems to monitor the status of the other of said active and standby systems,

wherein said each of said active and standby systems monitors, via said cluster network, the status of the other of said active and standby systems by transmitting status communications between said each of said active and standby systems and the other of said active and standby systems,

wherein said inter-storage network is used to conduct a remote copy function between said active and standby systems,

wherein said remote copy function operates, via said inter-storage network, such that when said active host computer writes data in said active storage system, a copy of said data is stored in said standby storage system, and

wherein the active wireless communication module, directly coupled to the active system, wirelessly transmits the second failure report to the standby wireless communication module, directly coupled to the standby system, when both of the cluster network and the inter-storage network have the network disconnection state at the same time.

15. A failover method for a cluster computing system including an active system having an active host computer and an active storage system, connected to said active host computer, for storing data accessible by said active host computer, a standby system having a standby host computer and a standby storage system, connected to said standby host computer, for storing data accessible by said standby host computer, a wireline network which connects said active and standby systems to each other, an active wireless communications module, directly coupled to the active system and a standby wireless communications module directly coupled to the standby system, each wireless communication module for performing wireless communications, the failover method comprising the steps of:

making the active wireless communication module, directly coupled to the active system, to detect a system condition of the active system;

when the active wireless communication module, directly coupled to the active system, detects a failure on the active system, making the active wireless communication module, directly coupled to the active system, to wirelessly communicate a first failure report to the standby wireless communication module at the standby system,

wherein said wireline network includes a cluster network which connects the host computers of the active and standby systems to each other and an inter-storage network which connects the storage systems of the active and standby systems to each other,

making the active wireless communication module, directly coupled to the active system, to detect whether a network disconnection state has occurred in said wireline network; and

when the active wireless communication module, directly coupled to the active system, detects that the network disconnection state has occurred in said wireline network, making the active wireless communication module, directly coupled to the active system, to wirelessly communicate a second failure report to the standby wireless communication module at the standby system,

wherein said cluster network is used by each of said active and standby systems to monitor the status of the other of said active and standby systems,

wherein said each of said active and standby systems monitors, via said cluster network, the status of the other of said active and standby systems by transmitting status communications between said each of said active and standby systems and the other of said active and standby systems,

wherein said inter-storage network is further used to conduct a remote copy function between said active and standby systems,

wherein said remote copy function operates, via said inter-storage network, such that when said active host computer writes data in said active storage system, a copy of said data is stored in said standby storage system,

wherein, upon receiving the first failure report from the active wireless communication module, directly coupled to the active system, the standby wireless communication module, directly coupled to the standby system, issues a first operation processing instruction command to the standby host computer at the standby system to start a first processing operation,

wherein, upon receiving the second failure report from the active wireless communication module, directly coupled to the active system, the standby wireless communication module, directly coupled to the standby system, issues a second operation processing instruction command to the standby host computer at the standby system to start a second processing operation,

wherein said first operation processing instruction command and said second operation processing instruction command are different from each other, and

wherein the active wireless communication module, directly coupled to the active system, wirelessly transmits the second failure report to the standby wireless communication module, directly coupled to the standby system, when both of the cluster network and the inter-storage network have the network disconnection state at the same time.