Centralized predictive routing using delay predictability measurements

1. A method, comprising:

receiving, at a device in a network, a routing strategy instruction that specifies a predictability threshold for communication delays in the network;

estimating, at the device, the communication delays for a plurality of paths in the network;

determining, by the device, predictability measurements for the estimated delays;

selecting, by the device, from among the plurality of paths, a particular path that has a predictability measurement that satisfies the predictability threshold and has a minimal estimated delay of the estimated communication delays that satisfy the predictability threshold, wherein the particular path is selected, based in part, on how predictable a corresponding estimated delay of the estimated communication delays is along the particular path, wherein the predictability of the estimated delay is quantified by a distribution function; and

installing, by the device, the particular path at one or more other devices in the network.

2. The method as in claim 1 , further comprising:

determining a level of confidence value for an estimated communication delay for one of the plurality of paths; and

determining that the level of confidence value is below a confidence threshold.

3. The method as in claim 2 , further comprising:

providing a notification that the level of confidence value is below the confidence threshold.

4. The method as in claim 2 , further comprising:

collecting metrics regarding the path associated with the level of confidence value that is below the confidence threshold; and

re-estimating the communication delays along the path associated with the level of confidence value.

5. The method as in claim 1 , wherein estimating the communication delays comprises:

generating delay probability distribution functions for the plurality of paths.

6. The method as in claim 5 , wherein determining the predictability measurements comprises:

calculating one or more metrics regarding the delay probability distribution functions as the predictability measurements.

7. The method as in claim 6 , wherein the one or more metrics regarding the delay probability distribution functions include an entropy measurement.

8. The method as in claim 6 , wherein the one or more metrics regarding the delay probability distribution functions include at least one of: a mean, a variance, a skewness metric, a kurtosis metric, or a Sarle's coefficient.

9. The method as in claim 1 , wherein the routing strategy instruction specifies a type of traffic for which the predictability threshold must be met.

10. An apparatus, comprising: one or more network interfaces to communicate with a low power and lossy network (LLN); a processor coupled to the network interfaces and adapted to execute one or more processes; and

a memory configured to store a process executable by the processor, the process when executed operable to:

receive a routing strategy instruction that specifies a predictability threshold for communication delays in the network;

estimate the communication delays for a plurality of paths in the network;

determine predictability measurements for the estimated delays;

select, from among the plurality of paths, a particular path that has a predictability measurement that satisfies the predictability threshold and has a minimal estimated delay of the estimated communication delays that satisfy the predictability threshold, wherein the particular path is selected, based in part, on how predictable a corresponding estimated delay of the estimated communication delays is along the particular path, wherein the predictability of the estimated delay is quantified by a distribution function; and

install the particular path at one or more devices in the network.

11. The apparatus as in claim 10 , wherein the process when executed is further operable to:

determine a level of confidence value for an estimated communication delay for one of the plurality of paths; and

determine that the level of confidence value is below a confidence threshold.

12. The apparatus as in claim 11 , wherein the process when executed is further operable to:

provide a notification that the level of confidence value is below the confidence threshold.

13. The apparatus as in claim 11 , wherein the process when executed is further operable to:

collect metrics regarding the path associated with the level of confidence value that is below the confidence threshold; and

re-estimate the communication delays along the path associated with the level of confidence value.

14. The apparatus as in claim 10 , wherein the communication delays are estimated by:

generating delay probability distribution functions for the plurality of paths.

15. The apparatus as in claim 14 , wherein the predictability measurements are estimated by:

calculating one or more metrics regarding the delay probability distribution functions as the predictability measurements.

16. The apparatus as in claim 15 , wherein the one or more metrics regarding the delay probability distribution functions include an entropy measurement.

17. The apparatus as in claim 15 , wherein the one or more metrics regarding the delay probability distribution functions include at least one of: a mean, a variance, a skewness metric, a kurtosis metric, or a Sarle's coefficient.

18. The apparatus as in claim 10 , wherein the routing strategy instruction specifies a type of traffic for which the predictability threshold must be met.

19. A tangible, non-transitory, computer-readable media having software encoded thereon, the software when executed by a processor operable to:

receive a routing strategy instruction that specifies a predictability threshold for communication delays in the network;

estimate the communication delays for a plurality of paths in the network;

determine predictability measurements for the estimated delays;

select, from among the plurality of paths, a particular path that has a predictability measurement that satisfies the predictability threshold and has a minimal estimated delay of the estimated communication delays that satisfy the predictability threshold, wherein the particular path is selected, based in part, on how predictable a corresponding estimated delay of the estimated communication delays is along the particular path, wherein the predictability of the estimated delay is quantified by a distribution function; and

install the particular path at one or more devices in the network.

20. The computer-readable media as in claim 19 , wherein the software when executed is further operable to:

determine a level of confidence value for an estimated communication delay for one of the plurality of paths; and

determine that the level of confidence value is below a confidence threshold.