US 2023/0412488 A1Published application

Link behavior prediction for use in path selection

AssigneeJuniper Networks, Inc.

InventorsSanjeev Kumar Mishra, Sabyasachi Mukhopadhyay, Shivaprasad Gali, Hsiuyen Tsai

Life of this patent

Estimating the term…

Loading patent…

Abstract

Techniques are described for predicting future behavior of links in a network and generating dynamic thresholds for link metrics for use in path selection. In one example, a computing system receives historical values of a link metric for links of a network. The computing system executes a machine learning system which processes the historical values of the link metric to generate: (1) a predicted future value of the link metric for each link; and (2) a threshold for the link metric indicating whether the predicted future value for each link is anomalous. The computing system computes a path based on the predicted future values of the link metric and the threshold for the link metric. The computing system provisions the computed path, thereby enabling a network device to forward network traffic along the computed path.

Claims20

Dependent claims

A method comprising:

receiving, by processing circuitry of a computing system, values of a link metric for each link of a plurality of links of a network over a historical time period;
processing, by a machine learning system executed by the processing circuitry, the values of the link metric for each link of the plurality of links to generate:
1) a predicted future value of the link metric for each link of the plurality of links;
and
2) a threshold for the link metric, the threshold indicative of whether a value of the link metric for each link of the plurality of links at a given time is anomalous;
determining, by the processing circuitry and based on the predicted future value of the link metric for each link of the plurality of links and the threshold, whether the predicted future value of the link metric for each link is anomalous;
computing, by the processing circuitry and based on the determination, a path comprising at least some of the plurality of links and excluding at least one link of the plurality of links determined to be anomalous;
and
provisioning, by the processing circuitry, the computed path to enable a network device to forward network traffic along the computed path.

6 more elements

A computing system comprising:

a storage device;
and
processing circuitry operably coupled to the storage device, the processing circuitry configured to:
receive values of a link metric for each link of a plurality of links of a network over a historical time period;
execute a machine learning system configured to process the values of the link metric for each link of the plurality of links to generate:
1) a predicted future value of the link metric for each link of the plurality of links;
and
2) a threshold for the link metric, the threshold indicative of whether a value of the link metric for each link of the plurality of links at a given time is anomalous;
determine, based on the predicted future value of the link metric for each link of the plurality of links and the threshold, whether the predicted future value of the link metric for each link is anomalous;
compute, based on the determination, a path comprising at least some of the plurality of links and excluding at least one link of the plurality of links determined to be anomalous;
and
provision the computed path to enable a network device to forward network traffic along the computed path.

9 more elements

Dependencies are inferred from the claim text.