Network system fault resolution via a machine learning model
AssigneeHewlett Packard Enterprise Development LP
InventorsJisheng Wang, Xiaoying Wu, Shmuel Shaffer, David Jea
Life of this patent
Estimating the term…
Abstract
Disclosed are embodiments for automatically resolving faults in a complex network system. Some embodiments monitor one or more of system operational parameter values and message exchanges between network components. A machine learning model detects a fault in the complex network system, and an action is selected based on a cause of the fault. After the action is applied to the complex network system, additional monitoring is performed to either determine the fault has been resolved or additional actions are to be applied to further resolve the fault.
Claims20
A network management system comprising processing circuitry configured to:
- determine, based on a plurality of candidate causes of a fault in operation of a network system and corresponding confidences in the plurality of candidate causes of the fault, a candidate cause of the fault from the plurality of candidate causes of the fault having a highest confidence of the corresponding confidences;
- identify a diagnostic action for diagnosing the fault;
- based on determining that a cost of the identified diagnostic action satisfies a diagnostic threshold, perform the diagnostic action to diagnose the fault to obtain a root cause of the fault, the diagnostic threshold configured based at least in part on the determined candidate cause;
- and
- based on determining that the cost of the identified diagnostic action does not satisfy the diagnostic threshold, perform a rectifying action to rectify the determined candidate cause, the rectifying action selected from a plurality of rectifying actions and having a highest confidence of rectifying the determined candidate cause.
A method comprising:
- determining, by a network management system comprising processing circuitry, based on a plurality of candidate causes of a fault in operation of a network system and corresponding confidences in the plurality of candidate causes of the fault, a candidate cause of the fault from the plurality of candidate causes of the fault having a highest confidence of the corresponding confidences;
- identifying, by the network management system, a diagnostic action for diagnosing the fault;
- based on determining that a cost of the identified diagnostic action satisfies a diagnostic threshold, performing, by the network management system, the diagnostic action to diagnose the fault to obtain a root cause of the fault, the diagnostic threshold configured based at least in part on the determined candidate cause;
- and
- based on determining that the cost of the identified diagnostic action does not satisfy the diagnostic threshold, performing, by the network management system, a rectifying action to rectify the determined candidate cause, the rectifying action selected from a plurality of rectifying actions and having a highest confidence of rectifying the determined candidate cause.
Non-transitory, computer-readable media comprising instructions that, when executed, are configured to cause processing circuitry of a network management system to:
- determine, based on a plurality of candidate causes of a fault in operation of a network system and corresponding confidences in the plurality of candidate causes of the fault, a candidate cause of the fault from the plurality of candidate causes of the fault having a highest confidence of the corresponding confidences;
- identify a diagnostic action for diagnosing the fault;
- based on determining that a cost of the identified diagnostic action satisfies a diagnostic threshold, perform the diagnostic action to diagnose the fault to obtain a root cause of the fault, the diagnostic threshold configured based at least in part on the determined candidate cause;
- and
- based on determining that the cost of the identified diagnostic action does not satisfy the diagnostic threshold, perform a rectifying action to rectify the determined candidate cause, the rectifying action selected from a plurality of rectifying actions and having a highest confidence of rectifying the determined candidate cause.
Dependencies are inferred from the claim text.