DETECTING ROOT CAUSE FOR TRANSACTION DEGRADATION USING CAUSAL BAYESIAN NETWORKS

• US 20150074035A1
• Filed: 09/02/2014
• Published: 03/12/2015
• Est. Priority Date: 09/02/2013
• Status: Active Grant

• Current Assignee: Appnomic Systems Pvt. Ltd.
• Original Assignee: Appnomic Systems Pvt. Ltd.
• Inventors: Narasappa, Sumanth
• Granted Patent: US 10,878,327 B2
• Time in Patent Office: Days
• US Class Current: 706/52
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Techniques for detecting root cause for transaction degradation using causal Bayesian networks are disclosed. In some embodiments, various states associated with an application comprising transactions and components are determined, wherein the determined states are associated with the application transactions and components. The determined states are used as input to build a Bayesian network whose nodes represent application transactions and components. A root cause set comprising one or more application components that is associated with a transaction degradation is inferred by traversing the Bayesian network.

Claims

1. A system, comprising:

a processor configured to;

determine states associated with an application, wherein the application comprises transactions and components and wherein the determined states are associated with the application transactions and components;

build a Bayesian network using the determined states as input, wherein each node of the Bayesian network corresponds to a transaction or component of the application;

traverse the Bayesian network to infer a root cause set associated with a transaction degradation, wherein the root cause set comprises one or more components of the application; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system of claim 1, wherein a transaction comprises a unit of work to achieve a is domain function of the application.

3. The system of claim 1, wherein a component comprises a resource that is used by a transaction to get work done.

4. The system of claim 1, wherein to determine states associated with the application comprises to measure metrics associated with application transactions and components and to map measured metrics to states based on thresholds.

5. The system of claim 1, wherein to determine states associated with the application comprises to determine states at prescribed polling intervals.

6. The system of claim 1, wherein to determine states associated with the application comprises to determine states using one or more anomaly detection algorithms.

7. The system of claim 1, wherein to traverse the Bayesian network to infer a root cause set associated with a transaction degradation comprises to identify related application components and dependent transactions.

8. The system of claim 1, wherein to traverse the Bayesian network to infer a root cause set associated with a transaction degradation comprises to detect the transaction degradation and include in the root cause set dependent components that are in anomalous states at one or more previous polling intervals.

9. The system of claim 1, wherein to traverse the Bayesian network to infer a root cause set associated with a transaction degradation comprises to query the built Bayesian network for the root cause set.

10. The system of claim 1, wherein the root cause set comprises components with anomalous KPIs (key performance indicators).

11. The system of claim 1, wherein the processor if further configured to receive feedback regarding the built Bayesian network from a domain expert.

12. The system of claim 1, wherein the processor if further configured to receive feedback regarding the inferred root cause set from a domain expert.

13. The system of claim 1, wherein the processor is further configured to update the built is Bayesian network based on operator feedback.

14. The system of claim 1, wherein the processor is further configured to update the built Bayesian network based on dynamic changes to the application.

15. The system of claim 1, wherein the Bayesian network comprises a causal network.

16. The system of claim 1, wherein the processor if further configured to predict impending transaction degradations and associated root cause sets using the built Bayesian network.

17. A method, comprising:

determining states associated with an application, wherein the application comprises transactions and components and wherein the determined states are associated with the application transactions and components;

building a Bayesian network using the determined states as input, wherein each node of the Bayesian network corresponds to a transaction or component of the application;

traversing the Bayesian network to infer a root cause set associated with a transaction degradation, wherein the root cause set comprises one or more components of the application.

18. The method of claim 17, further comprising adjusting the built Bayesian network based on operator feedback or application changes.

19. The method of claim 17, further comprising predicting impending transaction degradations and associated root cause sets using the built Bayesian network.

20. A computer program product embodied in a tangible computer readable storage medium and comprising computer instructions for:

determining states associated with an application, wherein the application comprises transactions and components and wherein the determined states are associated with the application transactions and components;

to building a Bayesian network using the determined states as input, wherein each node of the Bayesian network corresponds to a transaction or component of the application;

traversing the Bayesian network to infer a root cause set associated with a transaction degradation, wherein the root cause set comprises one or more components of the application.