Auto-scale performance assurance system and auto-scale performance assurance method
View Patent ↗An autoscale-type performance assurance system performs autoscaling to increase or reduce the number of VMs/containers V 1 to V 4 generated in a server and resources of V 1 to V 4 . A compute includes a plurality of types of V 1 to V 4 , a data collection unit that collects a resource allocation amount of V 1 to V 4 , and a resource control unit that performs autoscaling to increase or reduce the amount of resources of V 1 to V 4 according to a resource control amount. A controller includes a dependency calculation unit that calculates, based on the collected resource allocation amount, a degree of dependency indicating whether the resource allocation amount is dependent on a performance related to V 1 to V 4 for providing a communication service quality, and an autoscaling determination unit that obtains a resource control amount for increasing or reducing only resources related to the calculated degree of dependency indicating being dependent.
1. An autoscale-type performance assurance system configured to perform autoscaling to increase or reduce, through addition or deletion, a number of virtual machines (VMs)/containers, which are either or both of VMs and containers generated in a network-connected server and resources typified by a central processing unit (CPU) and a memory of each of the number of VMs/containers, the autoscale-type performance assurance system comprising:
a first server including:
a plurality of types of the number of VMs/containers,
a collection unit configured to collect a resource allocation amount of each of the number of VMs/containers, and
a control unit configured to perform autoscaling to increase or reduce resources of each of the number of VMs/containers; and
a second server including:
a calculation unit configured to, based on the resource allocation amount that is collected, calculate a degree of dependency indicating whether the resource allocation amount is dependent on a performance related to the number of VMs/containers for providing a communication service quality, and
a determination unit configured to obtain a resource control amount for increasing or reducing only resources related to the degree of dependency indicating being dependent that is calculated,
wherein the control unit increases or reduces the resources related to the degree of dependency indicating being dependent of the number of VMs/containers by executing autoscaling according to the resource control amount obtained by the determination unit.
2. The autoscale-type performance assurance system according to claim 1 ,
wherein the calculation unit obtains a correlation coefficient between the number of VMs/containers serving as the resource allocation amount and the performance and calculates the degree of dependency, as being dependent, when the correlation coefficient that is obtained exceeds a predetermined threshold value,
the determination unit obtains a resource control amount for increasing or reducing only the number of VMs/containers related to the degree of dependency indicating being dependent that is calculated, and
the control unit increases or reduces the number of VMs/containers by executing autoscaling according to the resource control amount.
3. The autoscale-type performance assurance system according to claim 1 ,
wherein the calculation unit obtains a correlation coefficient between a number of resource allocation sets of the VMs/containers serving as the resource allocation amount and the performance and calculates the degree of dependency, as being dependent, when the correlation coefficient that is obtained exceeds a predetermined threshold value,
the determination unit obtains a resource control amount for increasing or reducing only the number of resource allocation sets of the VMs/containers related to the degree of dependency indicating being dependent that is calculated, and
the control unit increases or reduces the number of resource allocation sets of the VMs/containers by executing autoscaling according to the resource control amount.
4. The autoscale-type performance assurance system according to claim 1 ,
wherein the calculation unit obtains a correlation coefficient between the number of VMs/containers serving as the resource allocation amount and the performance and calculates a weight constant for each of the number of VMs/containers according to the correlation coefficient that is obtained,
the determination unit determines a number of each of the number of VMs/containers to increase or reduce according to the weight constant of each of the number of VMs/containers that is calculated and obtains a resource control amount for increasing or reducing the number of each of the number of VMs/containers at a ratio at which the number of each of the number of VMs/containers is to be increased or reduced, and
the control unit increases or reduces the number of VMs/containers by executing autoscaling according to the resource control amount.
5. The autoscale-type performance assurance system according to claim 1 ,
wherein the calculation unit uses a number of resource allocation sets of the number of VMs/containers to obtains a correlation coefficient between the number of resource allocation sets and the performance and calculate a weight constant for each of the number of VMs/containers according to the correlation coefficient that is obtained,
the determination unit determines the number of each of the number of VMs/containers to increase or reduce according to the weight constant of each of the number of VMs/containers that is calculated and obtains a resource control amount for increasing or reducing the number of resource allocation sets for each of the number of VMs/containers at a ratio at which the number of each of the number of VMs/containers is to be increased or reduced, and
the control unit increases or reduces the number of resource allocation sets for each of the number of VMs/containers by executing autoscaling according to the resource control amount.
6. The autoscale-type performance assurance system according to claim 1 ,
wherein the calculation unit obtains a correlation coefficient between the number of VMs/containers serving as the resource allocation amount and the performance, defines the degree of dependency when the correlation coefficient that is obtained exceeds a threshold value that is predetermined as being dependent and the degree of dependency when the correlation coefficient that is obtained is equal to or less than the threshold value as being not dependent, and calculates a weight constant for each of the number of VMs/containers according to the correlation coefficient of each of the number of VMs/containers being dependent other than VMs/containers being not dependent,
the determination unit determines a number of each of the VMs/containers to increase or reduce according to the weight constant of each of the VMs/containers that is calculated and obtains a resource control amount for increasing or reducing the number of each of the number of VMs/containers at a ratio at which the number of each of the VMs/containers is to be increased or reduced, and
the control unit increases or reduces the number of VMs/containers by executing autoscaling according to the resource control amount.
7. The autoscale-type performance assurance system according to claim 1 ,
wherein the calculation unit obtains a correlation coefficient between a number of resource allocation sets of the VMs/containers serving as the resource allocation amount and the performance, defines the degree of dependency when the correlation coefficient that is obtained exceeds a threshold value that is predetermined as being dependent and the degree of dependency when the correlation coefficient that is obtained is equal to or less than the threshold value as being not dependent, and calculates a weight constant for each of the number of VMs/containers according to the correlation coefficient of each of the number of VMs/containers being dependent other than VMs/containers being not dependent,
the determination unit determines a number of each of the number of VMs/containers to increase or reduce according to the weight constant of each of the number of VMs/containers that is calculated and obtains a resource control amount for increasing or reducing the number of resource allocation sets for each of the number of VMs/containers at a ratio at which the number of each of the number of VMs/containers is to be increased or reduced, and
the control unit increases or reduces the number of resource allocation sets for each of the number of VMs/containers by executing autoscaling according to the resource control amount.
8. An autoscale-type performance assurance method of a system configured to perform autoscaling to increase or reduce, through addition or deletion, the number of virtual machines (VMs/containers, which are either or both of VMs and containers generated in a network-connected server and resources typified by a central processing unit (CPU) and a memory of each of the number of VMs/containers,
wherein the system includes a first server in which a plurality of types of the number of VMs/containers are generated and a second server connected to the first server, the method comprising:
by the first server, collecting a resource allocation amount of each of the number of VMs/containers,
by the second server, based on the resource allocation amount that is collected, calculating a degree of dependency indicating whether the resource allocation amount is dependent on a performance related to the number of VMs/containers for providing a communication service quality,
by the second server, obtaining a resource control amount for increasing or reducing only resources related to the degree of dependency indicating being dependent that is calculated, and
by the first server, increasing or reducing the resources related to the degree of dependency indicating being dependent of the number of VMs/containers by executing autoscaling according to the resource control amount.