METRIC DRIVEN HOLISTIC NETWORK MANAGEMENT SYSTEM

• US 20160065438A1
• Filed: 11/06/2015
• Published: 03/03/2016
• Est. Priority Date: 05/25/2004
• Status: Active Grant

• Current Assignee: University of Florida Research Foundation, Incorporated (State University System of Florida)
• Original Assignee: University of Florida Research Foundation, Incorporated (State University System of Florida)
• Inventors: BIRD, ROBERT BARRETT, SAXON, WILLIAM WARD
• Granted Patent: US 10,033,610 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: H04L 41/046 comprising network management agents or mobile agents therefor
• CPC Class Codes: H04L 41/0631 using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0894 Policy-based network configuration management
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level

Abstract

A metric driven method for managing a network can include the step of specifying a profile for a network event. The profile can include multiple discrete factors and can be associated with at least one event operation. Metrics can be received from multiple remotely located components residing within a network space. The metrics can be compared to the discrete factors, where each compared metric can correspond to one of the factors. A determination based on this comparison can indicate whether the network event has occurred. When the network event has occurred, the previously specified event operation associated with the network event can be automatically executed.

Claims

1. A metric driven method for managing a network, comprising:

gathering, by an input manager within a network space, network metrics from at least one network component within the network space;

determining, by a data analysis manager within the network space, that a specific network event has occurred based at least in part upon a comparison of a set of event factors to corresponding network metrics, the set of event factors defined in an event profile corresponding to the specific network event, and each of the set of event factors comprising a defined metric condition associated with the occurrence of the specific network event; and

implementing, by an output manager within the network space, a network action in response to an indication from the data analysis manager that the specific network event has occurred, the network action corresponding to the specific network event.

2. The method of claim 1, wherein the network metrics are gathered by a plurality of input managers, each of the plurality of input managers associated with a sub-network of the network space.

3. The method of claim 1, comprising storing, by a storage manager within the network space, the network metrics gathered by the input manager, the network metrics stored within one or more devices within a storage network of the network space.

4. The method of claim 3, wherein the network metrics are gathered by a plurality of input managers within the network space and communicated to the storage manager for storing.

5. The method of claim 1, wherein the determination that the specific network event has occurred is based at least in part upon a correlation of the defined metric conditions with the corresponding network metrics.

6. The method of claim 5, wherein each of the set of event factors in the event profile comprise a predominance characteristic that provides a weighting of the association between that defined metric condition and the occurrence of the specific network event.

7. The method of claim 6, wherein the correlation of the defined metric conditions with the corresponding network metrics is a weighted correlation based upon the predominance characteristics of the set of event factors.

8. The method of claim 5, wherein the data analysis manager generates a correlation value from the correlation of the defined metric conditions with the corresponding network metrics and determines the network action based at least in part upon the correlation value, wherein the indication from the data analysis manager triggers the network action.

9. The method of claim 1, comprising:

determining, by a central manager within the network space, that information in the event profile stored in a data store has been updated, the data store residing in a computing space external to the network space; and

communicating, by the central manager, the information to the data analysis manager.

10. A network management system, comprising:

one or more computing devices within a network space, the one or more computing devices comprising one or more processors and memory; and

a plurality of deployed kernels stored in the memory of the one or more computing devices, where execution of the plurality of deployed kernels by the one or more processors causes the one or more computing devices to;

gather network metrics from at least one network component within the network space;

determine that a specific network event has occurred based at least in part upon a comparison of a set of event factors to corresponding network metrics, the set of event factors defined in an event profile corresponding to the specific network event, and each of the set of event factors comprising a defined metric condition associated with the occurrence of the specific network event; and

automatically implement a network action in response to the determination that the specific network event has occurred, the network action corresponding to the specific network event.

11. The system of claim 10, wherein the determination that the specific network event has occurred is based at least in part upon a correlation of the defined metric conditions with the corresponding network metrics.

12. The system of claim 11, wherein each of the set of event factors in the event profile comprise a predominance characteristic that provides a weighting of the association between that defined metric condition and the occurrence of the specific network event.

13. The system of claim 12, wherein the correlation of the defined metric conditions with the corresponding network metrics is a weighted correlation based upon the predominance characteristics of the set of event factors.

14. The system of claim 11, wherein the network action is based at least in part upon a correlation value generated by the correlation of the defined metric conditions with the corresponding network metrics.

15. The system of claim 14, wherein the network action is selected from a plurality of network actions based upon a plurality of correlation levels, each of the plurality of correlation levels corresponding to at least one of the plurality of network actions.

16. The system of claim 10, wherein the plurality of deployed kernels comprises a data gathering kernel configured to gather at least a portion of the network metrics and a metric analyzing kernel configured to determine that the specific network event has occurred utilizing corresponding network metrics of the portion of the network metrics.

17. The system of claim 16, wherein the plurality of deployed kernels further comprises an execution kernel configured to implement the network action corresponding to the specific network event.

18. The system of claim 10, wherein execution of the plurality of deployed kernels by the one or more processors causes the one or more computing devices to:

detect a modification of a master kernel in a computing space external to the network space, the master kernel associated with at least one deployed kernel of the plurality of deployed kernels; and

automatically update, in response to the detection of the modification, the at least one deployed kernel to conform to the modification of the master kernel.

19. A metric driven method for managing a network, comprising:

deploying a plurality of software kernels into a network space, where the plurality of software kernels are configured to determine whether a network event has occurred based upon metrics of the network space and, in response to the determination, automatically execute a network action corresponding to the network event;

modifying a master kernel in a computing space external to the network space to produce a modified master kernel, the master kernel associated with at least one deployed kernel of the plurality of software kernels; and

responsive to the master kernel modification, automatically updating the at least one deployed kernel to conform to the modified master kernel.

20. The method of claim 19, comprising:

determining that a specific network event has occurred based at least in part upon a comparison of a set of event factors to corresponding network metrics of the network space, the set of event factors defined in an event profile corresponding to the specific network event, and each of the set of event factors comprising a defined metric condition associated with the occurrence of the specific network event; and

automatically implementing a network action in response to the determination that the specific network event has occurred, the network action corresponding to the specific network event.