Technique for autonomous network provisioning

1. A method for autonomous provisioning of a network comprising a plurality of nodes, the method comprising the steps of:

receiving at least one network-state measurement comprising at least one of a traffic measurement and a performance measurement, wherein the traffic measurement comprises constituent traffic, wherein the constituent traffic comprises a combination of a primary traffic, a secondary traffic, and a denied traffic;

determining at least one critical link based at least in part on the at least one network-state measurement, wherein the step of determining at least one critical link further comprises the steps of:

marking a traffic request according to a status of a selected route;

classifying at least one link according to at least one measurement of the marked traffic request; and

identifying at least one primary link having a ratio of primary traffic over secondary traffic above a predetermined threshold; and

formulating at least one link provisioning recommendation for the at least one critical link.

2. The method of claim 1 , wherein the performance measurement comprises at least one of a routing index and a resource allocation index, the routing index measuring efficacy of a routing function and the resource allocation index measuring efficacy of a resource allocation function.

3. The method of claim 1 , wherein the step of determining further comprises the step of:

determining whether denied traffic is above a predetermined denied traffic threshold.

4. The method of claim 1 , wherein the step of formulating further comprises the step of:

identifying the at least one critical link for capacity augmentation.

5. The method of claim 1 , wherein the step of formulating further comprises the step of:

calculating a capacity for the at least one critical link.

6. The method of claim 1 , wherein the step of formulating further comprises the step of:

implementing a learning method based on at least one historic network-state record.

7. The method of claim 6 , wherein the learning method determines inter-dependence among a combination of network performance, at least one traffic measurement and at least one resource capacity from a plurality of historic network-state records.

8. The method of claim 7 , wherein network performance comprises a quantification of at least one denied traffic request.

9. The method of claim 7 , wherein the at least one traffic measurement comprises constituent traffic, wherein constituent traffic comprises a combination of a primary traffic, a secondary traffic and a denied traffic.

10. The method of claim 7 , wherein the at least one resource capacity comprises at least one link capacity.

11. The method of claim 7 , wherein the learning method uses regression analysis based on at least two blocking vs. occupancy records for a link-type wherein the link-type is based on at least a link capacity and a blocking vs. occupancy record comprising a link performance measurement and a link capacity measurement.

12. The method of claim 7 , further comprising the step of:

determining an objective link occupancy yielding a projected blocking probability equal to or less than a blocking-probability target.

13. The method of claim 1 , further comprising the step of:

distributing a provisioning resource budget to the at least one critical link according to the at least one link provisioning recommendation.

14. The method of claim 2 , wherein the routing function is a source routing function and the step of determining further comprises the step of marking a traffic request according to a status of a selected route from a route set containing at least one route.

15. The method of claim 2 , wherein the routing function is a hop-by-hop routing function and the step of determining further comprises the step of marking a traffic request according to a status of a selected link from a link set containing at least one link.

16. At least one processor readable storage medium for storing a computer program of instructions configured to be readable by at least one processor for instructing the at least one processor to execute a computer process for performing the method as recited in claim 1 .

17. A system for autonomously provisioning a network, the system comprising:

a network controller for receiving at least one network-state measurement comprising at least one of a traffic measurement and a performance measurement, wherein the traffic measurement comprises constituent traffic, wherein the constituent traffic comprises a combination of primary traffic, secondary traffic, and denied traffic;

a determining module for determining at least one critical link based at least in part on the at least one network-state measurement, wherein the determining module further marks a traffic request according to a status of a selected route; classifies at least one link according to at least one measurement of the marked traffic request; and identifies at least one primary link having a ratio of primary traffic over secondary traffic above a predetermined threshold; and

a provisioning module for formulating at least one link provisioning recommendation for the at least one critical link.

18. The system of claim 17 , wherein the performance measurement comprises at least one of a routing index and a resource allocation index, the routing index measuring efficacy of a routing function and the resource allocation index measuring efficacy of a resource allocation function.

19. The system of claim 17 , wherein the determining module further determines whether denied traffic is above a predetermined denied traffic threshold.

20. The system of claim 17 , wherein the provisioning module further identifies the at least one critical link for augmentation.

21. The system of claim 17 , wherein the provisioning module further calculates a capacity for the at least one critical link.

22. The system of claim 21 , further comprising a budget module for apportioning a provisioning resource budget among the at least one critical link according to the at least one link provisioning recommendation.

23. The system of claim 17 , wherein a learning method based on at least one historic network-state record is implemented in formulating the at least one link provisioning recommendation.

24. The system of claim 23 wherein the learning method determines inter-dependence among a combination of network performance, at least one traffic measurement and at least one resource capacity from a plurality of historic network-state records.

25. The system of claim 24 , wherein network performance comprises a quantification of at least one denied traffic request.

26. The system of claim 24 , wherein the at least one traffic measurement comprises constituent traffic, wherein constituent traffic comprises a combination of a primary traffic, a secondary traffic and a denied traffic.

27. The system of claim 24 , wherein the at least one resource capacity comprises at least one link capacity.

28. The system of claim 24 , wherein the learning method uses regression analysis based on at least two blocking vs. occupancy records for a link-type wherein the link-type is based on at least a link capacity and a blocking vs. occupancy record comprising a link performance measurement and a link capacity measurement.

29. The system of claim 24 , wherein an objective link occupancy is determined to yield a projected blocking probability equal to or less than a blocking-probability target.

30. An article of manufacture for autonomously provisioning a network, the article of manufacture comprising:

at least one computer readable medium; and

instructions carried on the at least one computer readable medium;

wherein the instructions are configured to be readable from the at least one computer readable medium by at least one processor and thereby cause the at least one processor to operate so as to:

receive at least one network-state measurement comprising at least one of a traffic measurement and a performance measurement, wherein the traffic measurement comprises constituent traffic, wherein the constituent traffic comprises a combination of a primary traffic, a secondary traffic, and a denied traffic;

determine at least one critical link based at least in part on the at least one network-state measurement, wherein the step of determining at least one critical link further comprises the steps of:

marking a traffic request according to a status of a selected route;

classifying at least one link according to at least one measurement of the marked traffic request; and

identifying at least one primary link having a ratio of primary traffic over secondary traffic above a predetermined threshold; and

formulate at least one link provisioning recommendation for the at least one critical link.

31. The article of manufacturing of claim 30 , further causing the at least one processor to operate so as to:

identify the at least one critical link for augmentation; and

calculate a desired capacity for the at least one critical link.

32. A method for autonomous provisioning of a network, the method comprising the steps of:

receiving at least one network-state measurement comprising at least one of a traffic measurement and a performance measurement, wherein the traffic measurement comprises constituent traffic, wherein the constituent traffic comprises a combination of a primary traffic, a secondary traffic, and a denied traffic;

determining at least one critical link based at least in part on the at least one network-state measurement, wherein the step of determining at least on critical link further comprises the step of:

marking at least one traffic request according to a status of a selected route;

classifying at least one resource according to measurements of the at least one marked traffic request;

sending at least one measurement request to a node which follows a blocking node in a sequence of nodes selected for a traffic request; and

formulating at least one provisioning recommendation by learning from at least one historic network-state record.

33. The method of claim 32 wherein the network employs at least two protocols and the learning is based on regression analysis related to the at least two protocols.