Systems and methods for contextual network assurance based on change audits

Patent metadata

Field	Value
Publication number	US-2021075689-A1
Application number	US-201916562073-A
Country	US
Kind code	A1
Filing date	Sep 5, 2019
Priority date	Sep 5, 2019
Publication date	Mar 11, 2021
Grant date	—

Abstract

Aggregated health information for a managed network may be retrieved and processed in response to changes to the managed network topology, configuration, or software. In response to receiving notification that a change to a component of the managed network has occurred, a change audit analysis engine can retrieve performance indicator information from components along a traceroute including the component which underwent the change. The retrieved performance indicator information can be processed by a memory based neural network to predict an impact of the change on the aggregated health of the managed network. The predicted impact can be compared to network health information retrieved through an ongoing basis and issues can be determined based on a comparison of the predict impact and the retrieved health information.

First claim

What is claimed is:

1 . A method for identifying network infrastructure issues, the method comprising:
detecting a change to a network infrastructure, the change comprising one or more of a software modification or a configuration modification; receiving one or more performance indicator values from the network infrastructure; generating a predicted network performance based on the received one or more performance indicator values and a network snapshot comprising one or more preceding performance indicator values of the network infrastructure preceding the detected change, the predicted network performance generated by a neural network, the predicted network performance comprising one or more predicted performance indicator values; receiving one or more later performance indicator values from the network infrastructure; and determining a network infrastructure issue by comparing the predicted network performance with the received one or more later performance indicator values, wherein determining the network infrastructure issue comprises identifying a degree of deviation between the predicted network performance and the received one or more later performance indicator values, the identified degree of deviation exceeding a predetermined threshold.

2 . The method of claim 1 , wherein the neural network is a recurrent neural network (RNN).

3 . The method of claim 2 , wherein the RNN includes a memory component.

4 . The method of claim 3 , wherein the memory component is a long short-term memory (LSTM).

5 . The method of claim 1 , further comprising generating an alert indicating the determined network infrastructure issue, the alert comprising one or more of an interface alert, a text message, or an email.

6 . The method of claim 1 , further comprising generating a graphical user interface (GUI) comprising a network health trend line and one or more detected event bars, at least one of the one or more detected event bars corresponding to the detected change to the network infrastructure.

7 . The method of claim 6 , wherein the detected event bars are interactable and interacting with one of the detected event bars generates an information modal comprising summary information of the corresponding detected change and a respective impact on network performance.

8 . A system for identifying network infrastructure issues, the system comprising:
one or more processors; and a memory comprising instructions for the one or more processors to:
detect a change to a network infrastructure, the change comprising one or more of a software modification or a configuration modification;
receive one or more performance indicator values from the network infrastructure;
generate a predicted network performance based on the received one or more performance indicator values and a network snapshot comprising one or more preceding performance indicator values of the network infrastructure preceding the detected change, the predicted network performance generated by a neural network, the predicted network performance comprising one or more predicted performance indicator values;
receive one or more later performance indicator values from the network infrastructure; and
determine a network infrastructure issue by comparing the predicted network performance with the received one or more later performance indicator values, wherein determining the network infrastructure issue comprises identifying a degree of deviation between the predicted network performance and the received one or more later performance indicator values, the identified degree of deviation exceeding a predetermined threshold.

9 . The system of claim 8 , wherein the neural network is a recurrent neural network (RNN).

10 . The system of claim 9 , wherein the RNN includes a memory component.

11 . The system of claim 10 , wherein the memory component is a long short-term memory (LSTM).

12 . The system of claim 8 , wherein the memory further comprises instructions to generate an alert indicating the determined network infrastructure issue, the alert comprising one or more of an interface alert, a text message, or an email.

13 . The system of claim 8 , wherein the memory further comprises instructions to generate a graphical user interface (GUI) comprising a network health trend line and one or more detected event bars, at least one of the one or more detected event bars corresponding to the detected change to the network infrastructure.

14 . The system of claim 13 , wherein the detected event bars are interactable and interacting with one of the detected event bars generates an information modal comprising summary information of the corresponding detected change and a respective impact on network performance.

15 . A non-transitory computer readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to:
detect a change to a network infrastructure, the change comprising one or more of a software modification or a configuration modification; receive one or more performance indicator values from the network infrastructure; generate a predicted network performance based on the received one or more performance indicator values and a network snapshot comprising one or more preceding performance indicator values of the network infrastructure preceding the detected change, the predicted network performance generated by a neural network, the predicted network performance comprising one or more predicted performance indicator values; receive one or more later performance indicator values from the network infrastructure; and determine a network infrastructure issue by comparing the predicted network performance with the received one or more later performance indicator values, wherein determining the network infrastructure issue comprises identifying a degree of deviation between the predicted network performance and the received one or more later performance indicator values, the identified degree of deviation exceeding a predetermined threshold.

16 . The non-transitory computer readable medium of claim 15 , wherein the neural network is a recurrent neural network (RNN).

17 . The non-transitory computer readable medium of claim 16 , wherein the RNN includes a memory component.

18 . The non-transitory computer readable medium of claim 17 , wherein the memory component is a long short-term memory (LSTM).

19 . The non-transitory computer readable medium of claim 15 , further comprising instructions that cause the one or more processors to generate an alert indicating the determined network infrastructure issue, the alert comprising one or more of an interface alert, a text message, or an email.

20 . The non-transitory computer readable medium of claim 15 , further comprising instructions that cause the one or more processors to generate a graphical user interface (GUI) comprising a network health trend line and one or more detected event bars, at least one of the one or more detected event bars corresponding to the detected change to the network infrastructure, wherein the detected event bars are interactable and interacting with one of the detected event bars generates an information modal comprising summary information of the corresponding detected change and a respective impact on network performance.

Assignees

• Cisco Tech Inc

Inventors

• Ramanathan Shankar
• Natarajan Muhilan
• Salgueiro Gonzalo
• Henry Jerome
• Barton Robert Edgar

Classifications

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• H04L41/147 Primary

  for predicting network behaviour · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/09

  Supervised learning · CPC title

• H04L43/0817 Primary

  by checking functioning · CPC title