Data reporting method and related equipment

Abstract

The embodiment of the invention discloses a data reporting method and related equipment. The method can be used in the field of network monitoring. The method comprises the following steps: after network equipment (or network management equipment) obtains a sampling data set, training each model in a model set by using a part of sampling data in the sampling data set to obtain a value of a parameter included in each model; employing the other part of the sampling data in the sampling data set for verifying each model to obtain the model (namely the first model) with the minimum deviation corresponding to the model; enabling the network management equipment to predict the trend condition of the real monitoring data on the network equipment only through the first model and the values of the parameters included in the first model; enabling the network equipment to only report the identifier of the first model and the value of the parameter corresponding to the first model, so the data reporting amount of the network equipment is effectively reduced, the bandwidth pressure is reduced, and the data return bandwidth is saved.

Description

technical field [0001] The present application relates to the field of network monitoring, in particular to a data reporting method and related equipment. Background technique [0002] Network monitoring refers to the monitoring and control of network devices (such as computers, mobile phones, etc.) Traditional network monitoring methods mainly include the deficiencies in the following aspects: 1) network management equipment obtains monitoring data of network equipment through pull mode, so that a large number of network equipment cannot be monitored at the same time, which limits network growth; 2) acquisition of monitoring data The accuracy is at the minute level. To improve the accuracy, we can only increase the query frequency of the network equipment, but this will lead to a high utilization rate of the central processing unit (CPU) of the network equipment and affect the normal operation of the network equipment; 3) Due to the existence of network transmission delay,...

AI Technical Summary

Problems solved by technology

[0002] Network monitoring refers to the monitoring and control of network devices (such as computers, mobile phones, etc.) Traditional network monitoring methods mainly include the deficiencies in the following aspects: 1) network management equipment obtains monitoring data of network equipment through pull mode, so that a large number of network equipment cannot be monitored at the same time, which limits network growth; 2) acquisition of monitoring data The accuracy is at the minute level. To improve the accuracy, we can only increase the query frequency of the network equipment, but this will lead to a high utilization rate of the central processing unit (CPU) of the network equipment and affect the normal operation of the network equipment; 3) Due to the existence of network transmission delay, the accuracy of the obtained network device monitoring data will be affected

[0004] In the above-mentioned data reporting method, there are the following problems: the periodic reporting method has a large amount of reporting, which will occupy more CPU, memory and network bandwidth of the network device, thereby increasing the reporting burden; and the threshold reporting method only reports abnormal data. The network management equipment cannot obtain the change status of the data within the threshold of the network equipment

Images