Selective and dynamic application-centric network measurement infrastructure
In one embodiment, a device in a network receives data indicative of traffic characteristics of traffic associated with a particular application. The device identifies one or more paths in the network via which the traffic associated with the particular application was sent, based on the traffic characteristics. The device determines a probing schedule based on the traffic characteristics. The probing schedule simulates the traffic associated with the particular application. The device sends probes along the one or more identified paths according to the determined probing schedule.
1. A method comprising:
receiving, at a device in a network, data indicative of traffic characteristics of traffic associated with a particular application;
identifying, by the device, one or more paths in the network via which the traffic associated with the particular application was sent, based on the traffic characteristics;
determining, by the device, a probing schedule based on the traffic characteristics, wherein the probing schedule simulates the traffic associated with the particular application; and
sending, by the device, probes along the one or more identified paths according to the determined probing schedule,
receiving, at the device, result data regarding the sent probes;
determining, by the device, effects of the probes on the network based on the result data; and
adjusting, by the device, a probing strategy used by the device based on the effects of the probes on the network, wherein the probing strategy corresponds to at least one of: the probing schedule, the paths via which the probes are sent, or characteristics of the sent probes.
2. The method as in claim 1 , wherein the traffic characteristics comprise one or more of: packet size information regarding the traffic, flow duration information regarding the traffic, packet interval information regarding the traffic, or flow interval information regarding the traffic.
3. The method as in claim 1 , wherein the probes are sent in response to a determination that a performance metric for application-agnostic probes send along the one or more identified paths is below a threshold value.
4. The method as in claim 1 , wherein the effects of the probes correspond to a queue status or an available resource metric of the one or more nodes along the identified paths.
5. The method as in claim 4 , further comprising:
adjusting, by the device, the probing schedule to decrease a probing rate of the packets, based on the queue status indicating queuing delays or the available resource metric indicating that available resources of the one or more nodes are below a threshold level.
6. The method as in claim 1 , further comprising:
determining, by the device, a usefulness score based on the result data regarding the sent probes; and
adjusting, by the device, the probing strategy used by the device based in part on the usefulness score.
7. The method as in claim 1 , wherein the probing schedule is adjusted based on an identified transmission rate of the packets at which the one or more nodes begin experiencing delays.
8. The method as in claim 1 , further comprising:
providing, by the device, information obtained from the sent probes to a learning machine process.
9. An apparatus, comprising:
one or more network interfaces to communicate with a network;
a processor coupled to the one or more network interfaces and configured to execute a process; and
a memory configured to store the process executable by the processor, the process when executed operable to:
receive data indicative of traffic characteristics of traffic associated with a particular application;
identify one or more paths in the network via which the traffic associated with the particular application was sent, based on the traffic characteristics;
determine a probing schedule based on the traffic characteristics, wherein the probing schedule simulates the traffic associated with the particular application; and
send probes along the one or more identified paths according to the determined probing schedule,
receive result data regarding the sent probes;
determine effects of the probes on the network based on the result data; and
adjust a probing strategy used by the device based on the effects of the probes on the network, wherein the probing strategy corresponds to at least one of: the probing schedule, the paths via which the probes are sent, or characteristics of the sent probes.
10. The apparatus as in claim 9 , wherein the traffic characteristics comprise one or more of: packet size information regarding the traffic, flow duration information regarding the traffic, packet interval information regarding the traffic, or flow interval information regarding the traffic.
11. The apparatus as in claim 9 , wherein the probes are sent in response to a determination that a performance metric for application-agnostic probes send along the one or more identified paths is below a threshold value.
12. The apparatus as in claim 9 , wherein the effects of the probes correspond to a queue status or an available resource metric of the one or more nodes along the identified paths.
13. The apparatus as in claim 12 , wherein the process when executed is further operable to:
adjust the probing schedule to decrease a probing rate of the packets, based on the queue status indicating queuing delays or the available resource metric indicating that available resources of the one or more nodes are below a threshold level.
14. The apparatus as in claim 9 , wherein the process when executed is further operable to:
determine a usefulness score based on the result data regarding the sent probes; and
adjust the probing strategy, based in part on the usefulness score.
15. The apparatus as in claim 9 , wherein the probing schedule is adjusted based on an identified transmission rate of the packets at which the one or more nodes begin experiencing delays.
16. The apparatus as in claim 9 , wherein the process when executed is further operable to:
provide information obtained from the sent probes to a learning machine process.
17. A tangible, non-transitory, computer-readable media having software encoded thereon, the software when executed by a processor on a device in a computer network operable to:
receive data indicative of traffic characteristics of traffic associated with a particular application;
identify one or more paths in the network via which the traffic associated with the particular application was sent, based on the traffic characteristics;
determine a probing schedule based on the traffic characteristics, wherein the probing schedule simulates the traffic associated with the particular application; and
send probes along the one or more identified paths according to the determined probing schedule,
receive result data regarding the sent probes;
determine effects of the probes on the network based on the result data; and
adjust a probing strategy used by the device based on the effects of the probes on the network, wherein the probing strategy corresponds to at least one of: the probing schedule, the paths via which the probes are sent, or characteristics of the sent probes.