Impact-aware mitigation for computer networks

1. A method performed by a computing system, the method comprising:

for a computer network that includes a set of nodes in which each node is linked by a communication link to one or more other nodes of the set of nodes:

obtaining a service level objective for client-resource data flows traversing the computer network between client-side and resource-side nodes of the set of nodes;

obtaining indication of a failure event at a network location of the computer network;

identifying a mitigation action for the failure event;

determining an estimated impact to the service level objective for the mitigation action by applying simulated client-resource data flows to a network topology model of the computer network in combination with the mitigation action and the failure event; and

outputting the estimated impact to the service level objective.

2. The method of claim 1 , wherein the mitigation action identified for the failure event is one of a plurality of candidate mitigation actions; and

wherein the method further comprises identifying the plurality of candidate mitigation actions for the failure event from a set of mitigation actions.

3. The method of claim 2 , further comprising:

determining an estimated impact to the service level objective for each of the plurality of candidate mitigation actions by applying the simulated client-resource data flows to the network topology model of the computer network in combination with that candidate mitigation action and the failure event; and

outputting the estimated impact to the service level objective for each of the plurality of candidate mitigation actions.

4. The method of claim 2 , further comprising:

determining an estimated impact to the service level objective for each of the plurality of candidate mitigation actions by applying the simulated client-resource data flows to the network topology model of the computer network in combination with that candidate mitigation action and the failure event; and

identifying a target mitigation action from the plurality of candidate mitigation actions based on a comparison of the estimated impacts of the plurality of candidate mitigation actions, wherein the target mitigation action is the mitigation action identified for the failure event.

5. The method of claim 2 , further comprising:

determining an estimated impact to the service level objective for each of the plurality of candidate mitigation actions by applying the simulated client-resource data flows to the network topology model of the computer network in combination with that candidate mitigation action and the failure event;

identifying a target mitigation action from the plurality of candidate mitigation actions based on a comparison of the estimated impacts of the plurality of candidate mitigation actions, wherein the target mitigation action is the mitigation action identified for the failure event; and

outputting an identifier of the target mitigation action.

6. The method of claim 1 , wherein the service level objective includes one or more of:

a throughput of the client-resource data flows,

a flow completion time of the client-resource data flows.

7. The method of claim 1 , wherein obtaining the indication of the failure event includes obtaining a failure vector that identifies: (1) the network location of the failure event, (2) a failure type of the failure event, and (3) a magnitude of the failure event; and

wherein the estimated impact to the service level objective is determined for the mitigation action based on the failure vector applied to the network topology model.

8. The method of claim 1 , wherein the estimated impact to the service level objective includes an estimated impact to a distribution of the service level objective for the mitigation action.

9. The method of claim 1 , wherein the simulated client-resource data flows include a first subset of long-duration data flows and a second subset of short-duration data flows; and

wherein the simulated client-resource data flows have a time-varying data transmission rate that is defined by a transmission control protocol.

10. A computing system, comprising:

a logic machine; and

a data storage machine having instructions stored thereon executable by the logic machine to:

for a computer network that includes a set of nodes in which each node is linked by a communication link to one or more other nodes of the set of nodes:

obtain a service level objective for client-resource data flows traversing the computer network between client-side and resource-side nodes of the set of nodes;

obtain indication of a failure event at a network location of the computer network;

identify a mitigation action for the failure event;

determine an estimated impact to the service level objective for the mitigation action by applying simulated client-resource data flows to a network topology model of the computer network in combination with the mitigation action and the failure event; and

output the estimated impact to the service level objective.

11. The computing system of claim 10 , wherein the mitigation action identified for the failure event is one of a plurality of candidate mitigation actions; and

wherein the instructions are further executable by the logic machine to identify the plurality of candidate mitigation actions for the failure event from a set of mitigation actions.

12. The computing system of claim 11 , wherein the instructions are further executable by the logic machine to:

determine an estimated impact to the service level objective for each of the plurality of candidate mitigation actions by applying the simulated client-resource data flows to the network topology model of the computer network in combination with that candidate mitigation action and the failure event; and

output the estimated impact to the service level objective for each of the plurality of candidate mitigation actions.

13. The computing system of claim 11 , wherein the instructions are further executable by the logic machine to:

determine an estimated impact to the service level objective for each of the plurality of candidate mitigation actions by applying the simulated client-resource data flows to the network topology model of the computer network in combination with that candidate mitigation action and the failure event; and

identify a target mitigation action from the plurality of candidate mitigation actions based on a comparison of the estimated impacts of the plurality of candidate mitigation actions, wherein the target mitigation action is the mitigation action identified for the failure event.

14. The computing system of claim 11 , wherein the instructions are further executable by the logic machine to:

determine an estimated impact to the service level objective for each of the plurality of candidate mitigation actions by applying the simulated client-resource data flows to the network topology model of the computer network in combination with that candidate mitigation action and the failure event;

identify a target mitigation action from the plurality of candidate mitigation actions based on a comparison of the estimated impacts of the plurality of candidate mitigation actions, wherein the target mitigation action is the mitigation action identified for the failure event; and

output an identifier of the target mitigation action.

15. The computing system of claim 10 , wherein the service level objective includes one or more of:

a throughput of the client-resource data flows,

a flow completion time of the client-resource data flows.

16. The computing system of claim 10 , wherein the indication of the failure event includes a failure vector that identifies: (1) the network location of the failure event, (2) a failure type of the failure event, and (3) a magnitude of the failure event; and

wherein the estimated impact to the service level objective is determined for the mitigation action based on the failure vector applied to the network topology model.

17. The computing system of claim 10 , wherein the estimated impact to the service level objective includes an estimated impact to a distribution of the service level objective for the mitigation action.

18. The computing system of claim 10 , wherein the simulated client-resource data flows include a first subset of long-duration data flows and a second subset of short-duration data flows; and

wherein the simulated client-resource data flows have a time-varying data transmission rate that is defined by a transmission control protocol.

19. A computer network, comprising:

a set of nodes in which each node is linked by a communication link to one or more other nodes of the set of nodes; and

a computing system programmed with instructions executable by the computing system to:

obtain a service level objective for client-resource data flows traversing the computer network between client-side and resource-side nodes of the set of nodes;

obtain indication of a failure event at a network location of the computer network;

identify a mitigation action for the failure event;

determine an estimated impact to the service level objective for the mitigation action by applying simulated client-resource data flows to a network topology model of the computer network in combination with the mitigation action and the failure event; and

output the estimated impact to the service level objective.

20. The computer network of claim 19 , wherein the service level objective includes a throughput of the client-resource data flows or a flow completion time of the client-resource data flows; and

wherein the estimated impact to the service level objective is the lowest estimated impact to the service level objective among a plurality of candidate mitigation actions identified by the computing system for the failure event.