Technique for selecting a path computation element based on response time delay
A technique efficiently selects a path computation element (PCE) to compute a path between nodes of a computer network. The PCE selection technique is illustratively based on dynamic advertisements of the PCE's available path computation resources, namely a predictive response time (PRT). To that end, the novel technique enables one or more PCEs to dynamically send (advertise) their available path computation resources to one or more path computation clients (PCCs). In addition, the technique enables the PCC to efficiently select a PCE (or set of PCEs) to service a path computation request based upon those available resources.
1. An apparatus comprising:
a network interface configured to receive a computed path between nodes of a computer network from at least one path computation element (PCE); and
a processor configured to compute a path between the nodes of the computer network when a response is received from the at least one PCE indicating an inability to comply with a maximum response time (MRT).
2. The apparatus of claim 1 ,
wherein,
the response includes an indication of congestion, and
the processor is further configured to determine if an alternate PCE has less congestion than the at least one PCE.
3. The apparatus of claim 1 , wherein the network interface is further configured to send a message to the at least one PCE in order to clear a path computation request.
4. The apparatus of claim 1 , wherein the network interface is further configured to redirect at least one of a path computation request or future path computation requests to an alternate PCE.
5. The apparatus of claim 1 , wherein the processor is further configured to assign values to the MRT according to types of path computation requests.
6. The apparatus of claim 5 , wherein the processor is further configured to assign shorter MRTs to types with higher priorities.
7. The apparatus of claim 1 , wherein the network interface is further configured to: receive a predictive response time (PRT) from the at least one PCE to a path computation request; and distribute subsequent path computation requests proportionally among the at least one PCE based on the PRT of the at least one PCE.
8. A method comprising:
receiving a computed path between nodes of a computer network from at least one path computation element (PCE); and
computing a path between the nodes of the computer network when a response is received from the at least one PCE indicating an inability to comply with a maximum response time (MRT).
9. The method of claim 8 , further comprising:
determining if an alternate PCE has less congestion than the at least one PCE.
10. The method of claim 8 , further comprising:
sending a message to the at least one PCE in order to clear a path computation request.
11. The method of claim 8 , further comprising:
redirecting at least one of a path computation request or future path computation requests to an alternate PCE.
12. The method of claim 8 , further comprising:
assigning values to the MRT according to types of path computation requests.
13. The method of claim 12 , wherein assigning values to the MRT according to types of path computation requests further includes assigning shorter MRTs to types with higher priorities.
14. The method of claim 8 , further comprising:
receiving a predictive response time (PRT) from the at least one PCE to a path computation request; and
distributing subsequent path computation requests proportionally among the at least one PCE based on the PRT of the at least one PCE.
15. A non-transitory computer-readable storage medium having stored therein instructions which, when executed by a processor, cause the processor to perform operations comprising:
receiving a computed path between nodes of a computer network from at least one path computation element (PCE); and
computing a path between the nodes of a computer network when a response is received from the at least one PCE indicating an inability to comply with a maximum response time (MRT).
16. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the processor, result in operations further comprising determining if an alternate PCE has less congestion than the at least one PCE.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the processor, result in operations further comprising assigning values to the MRT according to types of path computation requests.
18. The non-transitory computer-readable storage medium of claim 17 , wherein assigning values to the MRT according to types of path computation requests further includes assigning shorter MRTs to types with higher priorities.
19. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the processor, result in operations further comprising selecting a preferred PCE based on a predictive response time (PRT) of the at least one PCE, the selected PCE to service a path computation request and compute the path between the nodes of the computer network.
20. The non-transitory computer-readable storage medium of claim 19 , wherein the instructions, when executed by the processor, result in operations further comprising determining subsequent path computation requests to be distributed proportionally among the at least one PCE based on the PRT of the at least one PCE.