Network performance guarantee system and network performance guarantee method

摘要

PROBLEM TO BE SOLVED: To appropriately control a resource allocation amount such as VM/number of containers by auto scaling. A network performance guarantee system 10 performs auto scaling to increase or decrease the number of VMs/containers V1 to V4 (V1 to V4) and resources of V1 to V4 generated in a server. The data collection unit 11a collects the measurement data including the resource usage amount related to the resource operation according to the resource allocation amount (allocation amount) of V1 to V4 and the performance value of the communication service related to V1 to V4. From the performance values in the collected measurement data by the learning unit 12b, the performance value having a high correlation with the assigned amount is set as the model performance value. The estimated optimum calculation unit 12c obtains an estimated value of the performance according to the allocation change by regression analysis of the model performance value and the performance value of the resource corresponding to the model performance value, and the estimated value satisfies the target value of the performance value. In addition, the allocation amount that minimizes the allocation amount is calculated. The resource control unit 11b increases or decreases the resources of V1 to V4 by auto scaling according to the calculated allocation amount. [Selection diagram] Figure 1