Remediating anomalies in a self-healing network
Some embodiments of the invention provide a method of remediating anomalies in an SD-WAN implemented by multiple forwarding elements (FEs) located at multiple sites connected by the SD-WAN. The method determines that a particular anomaly detected in the SD-WAN requires remediation to improve performance for a set of one or more flows traversing through the SD-WAN. The method identifies a set of two or more remedial actions for remediating the particular anomaly in the SD-WAN. For each identified remedial action in the set, the method selectively implements the identified remedial action for a subset of the set of flows for a duration of time in order to collect a set of performance metrics associated with SD-WAN performance during the duration of time for which the identified remedial action is implemented. Based on the collected sets of performance metrics, the method uses a machine-trained process to select one of the identified remedial actions as an optimal remedial action in the set to implement for all of the flows in the set of flows. The method implements the selected remedial action for all of the flows in the set of flows.
1. A method of remediating anomalies in an SD-WAN (software-defined wide-area network) implemented by a plurality of forwarding elements (FEs) located at a plurality of sites connected by the SD-WAN, the method comprising:
determining that a particular anomaly detected in the SD-WAN requires remediation to improve performance for a set of one or more flows traversing through the SD-WAN;
identifying a set of two or more remedial actions for remediating the particular anomaly in the SD-WAN;
for each identified remedial action in the set, selectively implementing the identified remedial action for a subset of the set of flows for a duration of time in order to collect a set of performance metrics associated with SD-WAN performance during the duration of time for which the identified remedial action is implemented;
based on the collected sets of performance metrics, using a machine-trained process to select one of the identified remedial actions as an optimal remedial action in the set to implement for all of the flows in the set of flows; and
implementing the selected remedial action for all of the flows in the set of flows.
2. The method of claim 1 , wherein selectively implementing each identified remedial action in the set to collect a set of performance metrics for each remedial action further comprises using each set of collected performance metrics to compute a performance score for each remedial action.
3. The method of claim 1 , wherein the optimal remedial action comprises a remedial action having an optimal set of performance metrics compared to the set of performance metrics for each other remedial action in the set.
4. The method of claim 1 , wherein the set of flows comprises flows associated with a single application.
5. The method of claim 4 , wherein the set of flows associated with the single application comprises a first set of flows associated with a first application, wherein the set of remedial actions comprises remedial actions that affect the first set of flows and not a second set of traffic flows associated with a second application.
6. The method of claim 1 , wherein the set of flows comprises flows associated with a set of two or more applications.
7. The method of claim 6 , wherein each subset of flows comprise a subset of flows to or from at least one application in the set of applications.
8. The method of claim 1 , wherein the particular anomaly comprises an increased latency associated with a particular transit FE that forwards application data traffic between (i) one or more edge FEs located at one or more branch sites connected by the SD-WAN and (ii) one or more applications deployed to a particular cloud datacenter connected by the SD-WAN.
9. The method of claim 8 , wherein the particular transit FE comprises a first transit FE and the particular cloud datacenter comprises a first cloud datacenter connected to the SD-WAN, wherein the set of two or more remedial actions comprises at least (i) a first alternate route between the one or more edge FEs and the one or more applications deployed to a second cloud datacenter connected to the SD-WAN via a second transit FE, and (ii) a second alternate route between the one or more edge FEs and the one or more applications deployed to a third cloud datacenter connected to the SD-WAN via a third transit FE.
10. The method of claim 9 , wherein selectively implementing each remedial action for a subset of the set of flows for the duration of time comprises:
directing the one or more edge FEs to use the second transit FE to forward a first subset of flows to the one or more applications deployed to the second cloud datacenter; and
directing the one or more edge FEs to use the third transit FE to forwards a second subset of flows to the one or more applications deployed to the third cloud datacenter.
11. The method of claim 10 , wherein the one or more edge FEs use the first transit FE to forward a remaining third subset of flows to the one or more applications deployed to the first cloud datacenter.
12. The method of claim 10 , wherein said directing comprises (i) updating a primary transit FE assigned to the one or more edge FEs for the first subset of flows from the first transit FE to the second transit FE and (ii) updating a primary transit FE assigned to the one or more edge FEs for the second subset of flows from the first transit FE to the third transit FE.
13. The method of claim 12 , wherein:
when a performance score generated for the first alternate route is higher than a performance score generated for the second alternate route, the machine-trained process selects the first alternate route to implement for the set of flows; and
when the performance score generated for the second alternate route is higher than the performance score generated for the first alternate route, the machine-trained process selects the second alternate route to implement for the set of flows.
14. The method of claim 1 , wherein each respective subset of the set of flows is assigned to each identified remedial action in the set in a round-robin fashion.
15. A non-transitory machine readable medium storing a program for execution by a set of processing units, the program for remediating anomalies in an SD-WAN (software-defined wide-area network) implemented by a plurality of forwarding elements (FEs) located at a plurality of sites connected by the SD-WAN, the program comprising sets of instructions for:
determining that a particular anomaly detected in the SD-WAN requires remediation to improve performance for a set of one or more flows traversing through the SD-WAN;
identifying a set of two or more remedial actions for remediating the particular anomaly in the SD-WAN;
for each identified remedial action in the set, selectively implementing the identified remedial action for a subset of the set of flows for a duration of time in order to collect a set of performance metrics associated with SD-WAN performance during the duration of time for which the identified remedial action is implemented;
based on the collected sets of performance metrics, using a machine-trained process to select one of the identified remedial actions as an optimal remedial action in the set to implement for all of the flows in the set of flows; and
implementing the selected remedial action for all of the flows in the set of flows.
16. The non-transitory machine readable medium of claim 15 , wherein the set of instructions for selectively implementing each identified remedial action in the set to collect a set of performance metrics for each remedial action further comprises a set of instructions for using each set of collected performance metrics to compute a performance score for each remedial action.
17. The non-transitory machine readable medium of claim 15 , wherein the optimal remedial action comprises a remedial action having an optimal set of performance metrics compared to the set of performance metrics for each other remedial action in the set.
18. The non-transitory machine readable medium of claim 15 , wherein the set of flows comprises flows associated with a single application.
19. The non-transitory machine readable medium of claim 18 , wherein the set of flows associated with the single application comprises a first set of flows associated with a first application, wherein the set of remedial actions comprises remedial actions that affect the first set of flows and not a second set of traffic flows associated with a second application.
20. The non-transitory machine readable medium of claim 15 , wherein:
the set of flows comprises flows associated with a set of two or more applications; and
each subset of flows comprises a subset of flows to or from at least one application in the set of applications.