Computer system and scale-out method of computer system

1. A computer system comprising:

a data processing system that is built in a specific environment,

the data processing system including

a management node that accepts data processing,

a processing cluster that includes a plurality of processing nodes configured to perform, in a distributed manner, the data processing accepted by the management node, and

a storage that includes a first data storage area in which data necessary for the data processing is stored,

the data processing system being configured to be capable of executing scaling for increasing or decreasing the number of the processing nodes of the plurality of processing nodes that are resources for executing the data processing in the distributed manner, according to a load of the processing cluster, wherein

the data processing system,

in a case where the resources of the data processing system become insufficient due to an increase in the load of the processing cluster,

includes the processing nodes of the plurality of processing nodes the number of which can be increased or decreased and a storage cluster having a plurality of storage nodes the number of which can be increased or decreased, the data processing system being configured to be capable of executing scale-out for increasing the number of the processing nodes of the plurality of processing nodes that are the resources for executing the data processing in the distributed manner, in a cloud environment different from the specific environment that can communicate with the data processing system via a network, and

the data processing system,

in a case where the scale-out is executed in the cloud environment,

starts a data copy process of copying data that is stored in the first data storage area and that is to be used in the data processing, from the first data storage area of the data processing system to a second data storage area of the storage cluster via the network, and

executes, during a period of time from the start of the data copy process to an end of the data copy process, the scale-out in the cloud environment by increasing the number of the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the first data storage area of the storage via the network.

2. The computer system according to claim 1 , wherein

the specific environment is an on-premises environment, and

the cloud environment is a public cloud environment.

3. The computer system according to claim 1 , wherein the data processing system, during the period of time from the start of the data copy process to the end of the data copy process,

first executes the scale-out by increasing the number of the processing nodes of the plurality of processing nodes having no cache, and

in a case where a communication bandwidth of the network becomes congested after the scale-out due to the data copy process and the access made by the processing nodes of the plurality of processing nodes to the first data storage area via the network,

switches the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the first data storage area via the network from the processing nodes of the plurality of processing nodes having no cache to the processing nodes of the plurality of processing nodes that have caches, that execute the data processing in the distributed manner while accessing the data stored in the first data storage area of the storage via the network, and that cache a part of data transferred from the storage.

4. The computer system according to claim 3 , wherein the data processing system, during the period of time from the start of the data copy process to the end of the data copy process,

executes the scale-out by using the processing nodes of the plurality of processing nodes having no cache before the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner are switched from the processing nodes of the plurality of processing nodes having no cache to the processing nodes of the plurality of processing nodes having caches, and

executes the scale-out by using the processing nodes of the plurality of processing nodes having caches after the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner are switched from the processing nodes of the plurality of processing nodes having no cache to the processing nodes of the plurality of processing nodes having caches.

5. The computer system according to claim 1 ,

wherein the data processing system,

in a case where the data copy process is completed, switches the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the first data storage area via the network, to the processing nodes of the plurality of processing nodes that access the data stored in the second data storage area of the storage cluster, and executes the data processing in the distributed manner by the switched processing nodes of the plurality of processing nodes.

6. The computer system according to claim 5 , wherein

the data processing system,

in the case where the data copy process is completed,

executes the scale-out in the cloud environment by increasing the number of the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the second data storage area of the storage cluster, according to the load of the processing cluster.

7. The computer system according to claim 1 , wherein

the data processing system,

in a case where the data copy process is completed,

switches the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the first data storage area via the network, to the processing nodes of the plurality of processing nodes that have no cache and that access the data stored in the second data storage area of the storage cluster, and executes the data processing in the distributed manner by the switched processing nodes of the plurality of processing nodes that have no cache, and the data processing system,

in a case where the resources of the processing nodes of the plurality of processing nodes that have been subjected to the scale-out in the cloud environment are overloaded and the data communication bandwidth in the storage cluster is congested after the data copy process is completed,

increases the number of the storage nodes after the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the second data storage area are switched to the processing nodes of the plurality of processing nodes that have caches, that execute the data processing in the distributed manner while accessing the data stored in the second data storage area, and that cache a part of data transferred from the second data storage area.

8. The computer system according to claim 5 , wherein

the data processing system,

in a case where a state in which a load of the resources of the processing nodes of the plurality of processing nodes in the cloud environment is low and in which usage of the data communication bandwidth in the storage cluster is low has continued for a predetermined period of time or more,

executes scale-in for reducing the number of the processing nodes of the plurality of processing nodes in the cloud environment to 0, deletes the data in the storage cluster, and stops the storage cluster.

9. The computer system according to claim 1 , further comprising:

a display device connected to the data processing system,

wherein the display device displays a screen including system operation information of the data processing system.

10. A scale-out method of a computer system that is executed by a data processing system built in a specific environment,

the data processing system including

a management node that accepts data processing,

a processing cluster that includes a plurality of processing nodes configured to perform, in the distributed manner, the data processing accepted by the management node, and

a storage that includes a first data storage area in which data necessary for the data processing is stored,

the data processing system being configured to be capable of executing scaling for increasing or decreasing a number of the processing nodes of the plurality of processing nodes that are resources for executing the data processing in the distributed manner, according to a load of the processing cluster, wherein

the data processing system,

in a case where the resources of the data processing system become insufficient due to an increase in the load of the processing cluster,

includes the processing nodes of the plurality of processing nodes the number of which can be increased or decreased and a storage cluster having a plurality of storage nodes the number of which can be increased or decreased, and executes scale-out for increasing the number of the processing nodes of the plurality of processing nodes that are the resources for executing the data processing in the distributed manner, in a cloud environment different from the specific environment that can communicate with the data processing system via a network, and,

in a case where the scale-out is executed in the cloud environment,

starts a data copy process of copying data that is stored in the first data storage area and that is to be used in the data processing, from the first data storage area of the data processing system to a second data storage area of the storage cluster via the network, and

executes, during a period of time from the start of the data copy process to an end of the data copy process, the scale-out in the cloud environment by increasing the number of the processing nodes of the plurality of processing nodes that execute the data processing in the distributed manner while accessing the data stored in the first data storage area of the storage via the network.