PROACTIVE CONTROLLER FOR FAILURE RESILIENCY IN COMMUNICATION NETWORKS

• US 20150195190A1
• Filed: 12/02/2014
• Published: 07/09/2015
• Est. Priority Date: 12/02/2013
• Status: Active Grant

• Current Assignee: University Of Ontario Institute Of Technology
• Original Assignee: University Of Ontario Institute Of Technology
• Inventors: SHAH HEYDARI, SHAHRAM, Izaddoost, Alireza
• Granted Patent: US 9,590,892 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/004 Error avoidance G06F11/07 and subgroups take precedence
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/30 Monitoring
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet
• CPC Class Codes: H04L 41/0668 by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/149 for prediction of maintenance
• CPC Class Codes: H04L 45/22 Alternate routing
• CPC Class Codes: H04L 45/28 using route fault recovery
• CPC Class Codes: H04L 45/42 Centralised routing

Abstract

Network devices and systems relating to prevention of large-scale network failure and determining a probability of a large-scale network failure for a network. The network devices may control rerouting of network traffic from failed network paths to preventative paths with probabilities of failure below determined thresholds. The network devices monitor and process real time data for dynamic and proactive rerouting when large-scale network failure occurs or is likely to occur.

Claims

1. A network controller for a network of network paths defined by nodes and links comprising:

a network interface configured to monitor the network to detect large-scale real time network failure indicia to detect likelihood of a network failure involving one or more network paths; and

a routing control processor configured to determine, for each link and node of the network, a probability of failure, and, for each failed or likely to fail network path, determine a preventive protection path with a lower probability of failure based on the determined probabilities of failure for the links and nodes, and dynamically reroute network traffic for the failed or likely to fail network path to the preventive protection path.

2. The network controller of claim 1, wherein the routing control processor is further configured to:

calculate disaster parameters using data feeds from one or more sensors, and update a probabilistic failure model with the disaster parameters;

determine the probability of failure for each node or link of the network using the updated probabilistic failure model; and

determine a future risk of failure for each network path in the network using the updated probabilistic failure model and the probabilities of failure; and

mitigate failure impact by rerouting by based on the future risk of failure for each network path.

3. The network controller of claim 1, wherein the routing control processor is further configured to, for each network path determined to have already failed or likely to fail, calculate a shortest preventative path with a probability of failure less than a threshold limit, and reroute network traffic for the failed or likely to fail network path to the shortest preventative path.

4. The network controller of claim 3, wherein the shortest preventative path has a probability of failure less than an upper threshold limit.

5. The network controller of claim 3, wherein the shortest preventative path has a probability of failure less than an upper threshold limit and a lower threshold limit.

6. The network controller of claim 1, wherein the routing control processor is configured to determine an impact area using the detected large-scale real time network failure indicia, determine the probability of failure for each link or node of the network within the impact area, and reroute one or more network paths within the impact area to preventative paths.

7. The network controller of claim 1, wherein the routing control processor is configured to update the impact area as failure in a large-scale scenario expands over time, update the probability of failure for each link or node of the network within the updated impact area, and reroute the one or more network paths within the updated impact area to preventative paths.

8. The network controller of claim 1, wherein the routing control processor is configured to evaluate network performance by calculating a number of disrupted connections, a disruption time, and a rerouting time for a network failure event.

9. The network controller of claim 1, wherein the routing control processor is configured to compute, once network failure is detected or likely, a failure decay rate based on a level of intensity of the network failure and geographical characteristics for the network failure, and determine the probabilities of failure using the failure decay rate.

10. The network controller of claim 1, wherein the routing control processor is configured to actively and continuously calculate, for each network path in an impact radius, an end-to-end path failure probability using the probabilities of failure for the links and nodes, and reroute traffic for the network paths based on one or more threshold values and the end-to-end path failure probabilities, whether or not failure is predicted or detected.

11. The network controller of claim 1, wherein the routing control processor is configured to calculate for each source node and destination node in the network, a shortest preventative path and reroute network traffic prior to network failure detection.

12. A system for preventative network path rerouting comprising:

a network of network paths defined by nodes and links;

a plurality of sensors to monitor the network to collect large-scale real time network failure indicia; and

a network controller for the network comprising;

a network interface configured to monitor the plurality of sensors for the large-scale real time network failure indicia; and

a routing control processor configured to determine, for each link and node of the network, a probability of failure, and, for each failed or likely to fail network path, determine a preventive protection path with a lower probability of failure based on the determined probabilities of failure for the links and nodes, and dynamically reroute network traffic for the failed or likely to fail network path to the preventive protection path.

13. A disaster recovery controller in a network of nodes connected by links, wherein the controller is configured to execute the following process in a loop:

determine likelihood of occurrence of a large scale failure through communication with a network node or a sensor for disaster detection

compute network failure parameters for range of impact and speed of expansion based on failure indicia received from the network node or the sensor;

compute impact area using the network failure parameters to define a set of network paths for rerouting;

for each network path, rerouting data flow to a shortest preventative path having a probability of failure less than a threshold limit; and

continuously monitoring the network for further failure updates and re-execution of process.

14. The disaster recovery controller of claim 13, further configured to:

calculate disaster parameters using data feeds from received from the network node or the sensor, and update a probabilistic failure model with the disaster parameters and network failure parameters; and

determine the probability of failure for each node or link of the network using the updated probabilistic failure model.

15. The disaster recovery controller of claim 13, further configured to, for each network path for rerouting, calculate a shortest preventative path with a probability of failure less than a threshold limit, and reroute network traffic for the network path to the shortest preventative path.

16. The disaster recovery controller of claim 15, wherein the shortest preventative path has a probability of failure less than an upper threshold limit.

17. The disaster recovery controller of claim 15, wherein the shortest preventative path has a probability of failure less than an upper threshold limit and a lower threshold limit.

18. The disaster recovery controller of claim 13, further configured to determine the impact area using the detected large-scale real time network failure indicia, determine the probability of failure for each link or node of the network within the impact area, and reroute one or more network paths within the impact area to preventative paths.

19. The disaster recovery controller of claim 13, further configured to evaluate network performance by calculating a number of disrupted connections, a disruption time, and a rerouting time for a network failure event.

20. The disaster recovery controller of claim 13, further configured to compute, once network failure is detected or likely, a failure decay rate based on a level of intensity of the network failure and geographical characteristics for the network failure, and determine the probabilities of failure using the failure decay rate.