Predictive routing using risk and longevity metrics
In one embodiment, a device makes a prediction regarding service level agreement violations by a network transport available between a site and an online application. The device associates a risk metric with the prediction, based in part on a type of the network transport. The device computes a longevity metric for the prediction that indicates an expected validity period for the prediction. The device cause traffic to be routed between the site and the online application using the network transport, based on the prediction and its associated risk metric and its longevity metric.
1. A method comprising:
making, by a device, a prediction regarding a number of service level agreement violations by a network transport among a plurality of possible network transports available between a site and an online application;
associating, by the device and based on a type of the network transport, a risk metric with the prediction, wherein the risk metric is indicative of a risk that the prediction is unreliable;
computing, by the device, a longevity metric that indicates an expected amount of time that the prediction regarding the number of service level agreement violations by the network transport is expected to stay valid; and
causing, by the device, traffic to be routed between the site and the online application using the network transport, based on the prediction, its associated risk metric, and its longevity metric each falling within respective thresholds.
2. The method as in claim 1 , wherein causing the traffic to be routed between the site and the online application using the network transport comprises:
selecting the network transport from among the plurality of possible network transports between the site and the online application.
3. The method as in claim 1 , wherein the network transport comprises a metropolitan-area Ethernet connection or a Multiprotocol Label Switching (MPLS) connection, and its risk metric indicates a low risk of the prediction being unreliable.
4. The method as in claim 1 , wherein the network transport comprises a cellular connection and its risk metric indicates a high risk of the prediction being unreliable.
5. The method as in claim 1 , further comprising:
causing, by the device, the traffic to be routed using a different network transport, when performance metrics associated with the traffic deviate from the prediction by a threshold amount.
6. The method as in claim 5 , further comprising:
determining a root cause of the performance metrics associated with the traffic deviating from the prediction by a threshold amount.
7. The method as in claim 1 , wherein the device causes the traffic to be routed between the site and the online application using the network transport, based in part on the longevity metric exceeding a user-specified threshold.
8. The method as in claim 1 , wherein causing the traffic to be routed between the site and the online application using the network transport comprises:
providing, by the device, a routing recommendation to a user interface that recommends routing the traffic between the site and the online application using the network transport.
9. The method as in claim 8 , further comprising:
using, by the device, feedback from the user interface regarding the routing recommendation to control further routing recommendations.
10. The method as in claim 1 , wherein the online application is a software-as-a-service (SaaS) application.
11. An apparatus, comprising:
one or more network interfaces;
a processor coupled to the one or more network interfaces and configured to execute one or more processes; and
a memory configured to store a process that is executable by the processor, the process when executed configured to:
make a prediction regarding a number of service level agreement violations by a network transport among a plurality of possible network transports available between a site and an online application;
associate, based on a type of the network transport, a risk metric with the prediction, wherein the risk metric is indicative of a risk that the prediction is unreliable;
compute a longevity metric that indicates an expected amount of time that the prediction regarding the number of service level agreement violations by the network transport is expected to stay valid; and
cause traffic to be routed between the site and the online application using the network transport, based on the prediction, its associated risk metric, and its longevity metric each falling within respective thresholds.
12. The apparatus as in claim 11 , wherein the apparatus causes the traffic to be routed between the site and the online application using the network transport by:
selecting the network transport from among the plurality of possible network transports between the site and the online application.
13. The apparatus as in claim 11 , wherein the network transport comprises an Internet connection, and its risk metric indicates a medium risk of the prediction being unreliable.
14. The apparatus as in claim 11 , wherein the network transport comprises a cellular connection and its risk metric indicates a high risk of the prediction being unreliable.
15. The apparatus as in claim 11 , wherein the process when executed is further configured to:
cause the traffic to be routed using a different network transport, when performance metrics associated with the traffic deviate from the prediction by a threshold amount.
16. The apparatus as in claim 15 , wherein the process when executed is further configured to:
determine a root cause of the performance metrics associated with the traffic deviating from the prediction by a threshold amount.
17. The apparatus as in claim 11 , wherein the apparatus causes the traffic to be routed between the site and the online application using the network transport, based in part on the longevity metric exceeding a user-specified threshold.
18. The apparatus as in claim 11 , wherein the apparatus causes the traffic to be routed between the site and the online application using the network transport by:
providing a routing recommendation to a user interface that recommends routing the traffic between the site and the online application using the network transport.
19. The apparatus as in claim 18 , wherein the process when executed is further configured to:
use feedback from the user interface regarding the routing recommendation to control further routing recommendations.
20. A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:
making, by the device, a prediction regarding a number of service level agreement violations by a network transport among a plurality of possible network transports available between a site and an online application;
associating, by the device and based on a type of the network transport, a risk metric with the prediction, wherein the risk metric is indicative of a risk that the prediction is unreliable;
computing, by the device, a longevity metric that indicates an expected amount of time that the prediction regarding the number of service level agreement violations by the network transport is expected to stay valid; and
causing, by the device, traffic to be routed between the site and the online application using the network transport, based on the prediction, its associated risk metric, and its longevity metric each falling within respective thresholds.