Fault tolerant server

1. A fault tolerant server configured to duplicate information processing by an online subsystem including a processor and a memory and an offline subsystem including a processor and a memory, the fault tolerant server comprising:

an entire copy unit for executing entire copy processing for copying all data being stored in the memory of the online subsystem into the memory of the offline subsystem without stopping execution of information processing by the processor of the online subsystem, before start of duplication of the information processing by the online subsystem and the offline sub system;

a data detection unit for detecting data, the data satisfying a criterion indicating that content of data is changed during the entire copy processing, among data being stored in the memory of the online subsystem; and

a change data copy unit for copying the detected data from the memory of the online subsystem into the memory of the offline subsystem,

wherein the change data copy unit copies the detected data into the memory of the offline subsystem after finishing the entire copy processing,

wherein the change data copy unit stops the information processing by the processor of the online subsystem after finishing the entire copy processing, and executes batch copy processing for copying all the detected data from the memory of the online subsystem into the memory of the offline subsystem,

wherein after finishing the batch copy processing, the change data copy unit resumes the information processing by the processor of the online subsystem being stopped, also starts information processing by the processor of the offline subsystem, and starts duplication of information processing,

wherein after the batch copy processing and also after starting the duplication of information processing by the online subsystem and the offline subsystem, the change data copy unit executes undetected data copy processing for copying data being not detected as the data satisfying the criterion indicating that content of data is changed during the entire copy processing among the data being stored in the memory of the online subsystem, without stopping the duplication of the information processing by the online subsystem and the offline subsystem, and

wherein:

on an occasion of the undetected data copy processing, the change data copy unit checks whether or not data stored in the memory of the online subsystem to be copied is coincident with data stored in the memory of the offline subsystem, and as a result of checking,

in a case that the each data is coincident with each other, does not copy the data being stored in the memory of the online subsystem into the memory of the offline subsystem, and

in a case that the each data is not coincident with each other, copies the data being stored in the memory of online subsystem into respective the memory of the online subsystem and the memory of the offline subsystem.

2. The fault tolerant server according to claim 1 , wherein;

during the undetected data copy processing, the change data copy unit monitors data output respectively from each the processor of the online subsystem and the processor of the offline subsystem, and as a result of monitoring,

in a case that the each data is not coincident with each other, stops the duplication of the information processing by the online subsystem and the offline subsystem, and

copies collectively all the data being stored in the memory of the online subsystem into the memory of the offline subsystem.

3. The fault tolerant server according to claim 1 , wherein;

the data detection unit

calculates a summary of all data being stored in the memory of the online subsystem before the entire copy processing as a first summary,

calculates a summary of all data being stored in the memory of the online subsystem during the entire copy processing as a second summary, and

detects data of which the first summary and the second summary are different as the data satisfying the criterion indicating that content of data is changed during the entire copy processing.

4. The fault tolerant server according to claim 1 , wherein the data is not copied by the entire copy unit to a storage device other than the memory of the offline subsystem and the memory of the online subsystem.

5. The fault tolerant server according to claim 1 , wherein the entire copy processing is performed through a background processing without stopping the operation of an operating system of the fault tolerant server,

wherein the data detection unit performs a first-time checksum calculation prior to the entire copy processing, and subsequently, after starting the entire copy processing by the entire copy unit, and

wherein the entire copy unit performs the entire copy processing for copying all pages in the memory of the online subsystem into the offline subsystem after finishing of the first-time checksum calculation by the data detection unit.

6. A non-transitory computer readable medium storing a program comprising instructions for causing a control device, the control device being equipped in a fault tolerant server configured to duplicate information processing by an online subsystem including a processor and a memory and an offline subsystem including a processor and a memory, to function as:

an entire copy unit for executing entire copy processing for copying all data being stored in the memory of the online subsystem into the memory of the offline subsystem without stopping execution of information processing by the processor of the online subsystem, before start of duplication of the information processing by the online subsystem and the offline sub system;

a data detection unit for detecting data, the data satisfying a criterion indicating that content of data is changed during the entire copy processing, among data being stored in the memory of the online subsystem; and

a change data copy unit for copying the detected data from the memory of the online subsystem into the memory of the offline subsystem,

wherein the change data copy unit copies the detected data into the memory of the offline subsystem after finishing the entire copy processing,

wherein the change data copy unit stops the information processing by the processor of the online subsystem after finishing the entire copy processing, and executes batch copy processing for copying all the detected data from the memory of the online subsystem into the memory of the offline subsystem,

wherein after finishing the batch copy processing, the change data copy unit resumes the information processing by the processor of the online subsystem being stopped, also starts information processing by the processor of the offline subsystem, and starts duplication of information processing,

wherein after the batch copy processing and also after starting the duplication of information processing by the online subsystem and the offline subsystem, the change data copy unit executes undetected data copy processing for copying data being not detected as the data satisfying the criterion indicating that content of data is changed during the entire copy processing among the data being stored in the memory of the online subsystem, without stopping the duplication of the information processing by the online subsystem and the offline subsystem, and

wherein:

on an occasion of the undetected data copy processing, the change data copy unit checks whether or not data stored in the memory of the online subsystem to be copied is coincident with data stored in the memory of the offline subsystem, and as a result of checking,

in a case that the each data is coincident with each other, does not copy the data being stored in the memory of the online subsystem into the memory of the offline subsystem, and

in a case that the each data is not coincident with each other, copies the data being stored in the memory of online subsystem into respective the memory of the online subsystem and the memory of the offline subsystem.

7. The non-transitory computer readable medium storing the program according to claim 6 , wherein the data is not copied by the entire copy unit to a storage device other than the memory of the offline subsystem and the memory of the online subsystem constituting a duplex system.

8. The non-transitory computer readable medium storing the program according to claim 6 , wherein the entire copy processing is performed through a background processing without stopping the operation of an operating system of the fault tolerant server,

wherein the data detection unit performs a first-time checksum calculation prior to the entire copy processing, and subsequently, after starting the entire copy processing by the entire copy unit, and

wherein the entire copy unit performs the entire copy processing for copying all pages in the memory of the online subsystem into the offline subsystem after finishing of the first-time checksum calculation by the data detection unit.

9. A memory data copy method for a fault tolerant server configured to duplicate information processing by an online subsystem including a processor and a memory and an offline subsystem including a processor and a memory, the method comprising:

executing entire copy processing for copying all data being stored in the memory of the online subsystem into the memory of the offline subsystem without stopping execution of information processing by the processor of the online subsystem, before start of duplication of the information processing by the online subsystem and the offline subsystem;

detecting data, the data satisfying a criterion indicating that content of data is changed during the entire copy processing, among data being stored in the memory of the online subsystem;

copying the detected data from the memory of the online subsystem into the memory of the offline subsystem;

copying the detected data into the memory of the offline subsystem after finishing the entire copy processing;

stopping the information processing by the processor of the online subsystem after finishing the entire copy processing;

executing batch copy processing for copying all the detected data from the memory of the online subsystem into the memory of the offline subsystem;

after finishing the batch copy processing, resuming the information processing by the processor of the online subsystem being stopped, also starts information processing by the processor of the offline subsystem, and starts duplication of information processing;

after the batch copy processing and also after starting the duplication of information processing by the online subsystem and the offline subsystem, executing undetected data copy processing for copying data being not detected as the data satisfying the criterion indicating that content of data is changed during the entire copy processing among the data being stored in the memory of the online subsystem, without stopping the duplication of the information processing by the online subsystem and the offline subsystem;

on an occasion of the undetected data copy processing, checking whether or not data stored in the memory of the online subsystem to be copied is coincident with data stored in the memory of the offline subsystem, and as a result of checking;

in a case that the each data is coincident with each other, not copying the data being stored in the memory of the online subsystem into the memory of the offline subsystem; and

in a case that the each data is not coincident with each other, copying the data being stored in the memory of online subsystem into respective the memory of the online subsystem and the memory of the offline subsystem.

10. The memory data copy method according to claim 9 , wherein the data is not copied to a storage device other than the memory of the offline subsystem and the memory of the online subsystem constituting a duplex system.

11. The memory data copy method according to claim 9 , wherein the entire copy processing is performed through a background processing without stopping the operation of an operating system of the fault tolerant server,

wherein a first-time checksum calculation is performed prior to the entire copy processing, and subsequently, after starting the entire copy processing, and

wherein the entire copy processing is performed for copying all pages in the memory of the online subsystem into the offline subsystem after finishing of the first-time checksum calculation.