Network performance prediction method and apparatus
First Claim
1. A method of predicting network performance, comprising:
- obtaining, by a network performance prediction apparatus, M pieces of performance parameter data of a link;
calculating, by the apparatus, a cloud model digital feature of network performance of the link according to the M pieces of performance parameter data;
calculating, by the apparatus, N pieces of predicted performance parameter data of the network performance of the link according to the cloud model digital feature; and
raising, by the apparatus, a prediction alarm that the link has a risk of exceeding a limit if there is data exceeding a preset upper limit value in the N pieces of predicted performance parameter data, wherein M and N are preset positive integers;
executing, by the apparatus, when the data in the M pieces of performance parameter data exceeds the preset upper limit value, or when the data in the N pieces of predicted performance parameter data exceeds the preset upper limit value, active-standby link switching if there is a standby link for the link;
wherein the cloud model digital feature of the network performance of the link comprises;
an expected value Ex, an entropy En and a hyper entropy He of the M pieces of network performance data; and
wherein the calculating a cloud model digital feature of network performance of the link according to the M pieces of performance parameter data comprises;
calculating a sample mean of the M pieces of performance parameter data by using
1 Assignment
0 Petitions
Accused Products
Abstract
The present application relates to a method and apparatus of predicting network performance, wherein the method includes: obtaining M performance parameter data of a link; calculating a cloud model digital feature of network performance of the link according to the M performance parameter data; calculating N predicted performance parameter data of the network performance of the link according to the cloud model digital feature; and raising a prediction alarm that the link has a risk of exceeding a limit if there is data exceeding a preset upper limit value in the N predicted performance parameter data, wherein M and N are preset positive integers. A change trend of network performance is analyzed according to real-time monitored data of the network performance of the link to predict future performance parameter data, and whether an alarm is raised and protection switching is performed is determined based on the prediction.
9 Citations
8 Claims
-
1. A method of predicting network performance, comprising:
-
obtaining, by a network performance prediction apparatus, M pieces of performance parameter data of a link; calculating, by the apparatus, a cloud model digital feature of network performance of the link according to the M pieces of performance parameter data; calculating, by the apparatus, N pieces of predicted performance parameter data of the network performance of the link according to the cloud model digital feature; and raising, by the apparatus, a prediction alarm that the link has a risk of exceeding a limit if there is data exceeding a preset upper limit value in the N pieces of predicted performance parameter data, wherein M and N are preset positive integers; executing, by the apparatus, when the data in the M pieces of performance parameter data exceeds the preset upper limit value, or when the data in the N pieces of predicted performance parameter data exceeds the preset upper limit value, active-standby link switching if there is a standby link for the link; wherein the cloud model digital feature of the network performance of the link comprises;
an expected value Ex, an entropy En and a hyper entropy He of the M pieces of network performance data; and
wherein the calculating a cloud model digital feature of network performance of the link according to the M pieces of performance parameter data comprises;calculating a sample mean of the M pieces of performance parameter data by using - View Dependent Claims (2, 3, 4)
-
-
5. A network performance prediction apparatus, comprising:
- a processor and a memory having program code stored therein, wherein the program code executed on the processor configures the apparatus to;
obtain M pieces of performance parameter data of a link; calculate a cloud model digital feature of network performance of the link according to the M pieces of performance parameter data; calculate N pieces of predicted performance parameter data of the network performance of the link according to the cloud model digital feature; determine whether there is data exceeding a preset upper limit value in the N pieces of predicted performance parameter data; raise a prediction alarm that the link has a risk of exceeding a limit if there is the data exceeding the preset upper limit value in the N pieces of predicted performance parameter data, wherein M and N are preset positive integers; wherein the code executed on the processor further configures the apparatus to execute, when the data in the M pieces of performance parameter data exceeds the preset upper limit value, or when the data in the N pieces of predicted performance parameter data exceeds the preset upper limit value, active-standby link switching if there is a standby link for the link; wherein the cloud model digital feature of the network performance of the link calculated by the first calculating module comprises;
an expected value Ex, an entropy En and a hyper entropy He of the M pieces of performance parameter data;the code executed on the processor configures the apparatus to calculate a sample mean of the M pieces of performance parameter data by using - View Dependent Claims (6, 7, 8)
- a processor and a memory having program code stored therein, wherein the program code executed on the processor configures the apparatus to;
Specification