End-to-end service quality using source-routed probes
The need to monitor real time network services has prompted service providers to use new measurement technologies, such as service-specific probes. A service-specific probe is an active probe that closely mimics the service traffic so that it receives the same treatment from the network as the actual service traffic. Service-specific probes are end-to-end and their deployment depends on solutions that address questions such as minimizing probe traffic, while still obtaining maximum coverage of all the links in the network. A polynomial-time probe-path computation algorithm is provided as well as a 2-approximate solution for merging probe paths when the number of probes exceeds a required bound k. The algorithms are evaluated using ISP topologies generated via Rocketfuel. For most topologies, it is possible to cover more than about 98% of the edges using just about 5% of the nodes as terminals.
1. A method for network monitoring, comprising:
receiving by a device a topology of a network, a set of selected edges in the network, and a set of terminals in the network;
determining a set of sourcerouted probes having a plurality of probe paths that originate and terminate from the set of terminal nodes, the sourcerouted probes crafted to emulate different network services, the probe paths covering the set of selected edges, while minimizing cost; and
deploying the set of sourcerouted probes to measure an end-to-end service quality of a specific service being monitored.
2. The method of claim 1 , wherein no sourcerouted probe in the set of sourcerouted probes exceeds a predetermined link capacity.
3. The method of claim 1 , wherein a total cost of the set of sourcerouted probes is minimized.
4. The method of claim 1 , wherein a maximum cost of each sourcerouted probe in the set of sourcerouted probes is minimized.
5. The method of claim 1 , wherein a total number of source routed probes in the set of sourcerouted probes is minimized.
6. The method of claim 1 , wherein the set of sourcerouted probes is deployed to monitor a voice over Internet protocol (VoIP) service.
7. A system for network monitoring, comprising:
a processor including an operating system and at least one storage device; and
a software component stored on a computer readable non-transitory medium, the software component when processed by a computer, adapt the operation of the computer to perform a method of network monitoring, including:
receiving a topology of a network, a set of selected edges in the network, and a set of terminals in the network;
determining a set of sourcerouted probes having a plurality of probe paths that originate and terminate from the set of terminal nodes, the sourcerouted probes crafted to emulate different network services, the probe paths covering the set of selected edges, while minimizing cost; and
deploying the set of sourcerouted probes to measure an end-to-end service quality of a specific service being monitored.
8. The system of claim 7 , wherein no sourcerouted probe in the set of sourcerouted probes exceeds a predetermined link capacity.
9. The system of claim 7 , wherein a total cost of the set of sourcerouted probes is minimized.
10. The system of claim 7 , wherein a maximum cost of each sourcerouted probe in the set of sourcerouted probes is minimized.
11. The system of claim 7 , wherein a total number of sourcerouted probes in the set of sourcerouted probes is minimized.
12. The system of claim 7 , wherein the set of sourcerouted probes is deployed to monitor a voice over Internet protocol (VoIP) service.
13. A non-transitory computer readable medium storing instructions which, when processed by a computer, adapt the operation of the computer to perform a method for network monitoring, the method comprising:
receiving a topology of a network, a set of selected edges in the network, and a set of terminals in the network;
determining a set of sourcerouted probes having a plurality of probe paths that originate and terminate from the set of terminal nodes, the sourcerouted probes crafted to emulate different network services, the probe paths covering the set of selected edges, while minimizing cost; and
deploying the set of sourcerouted probes to measure an end-to-end service quality of a specific service being monitored.
14. The non-transitory computer readable medium of claim 13 , wherein no sourcerouted probe in the set of sourcerouted probes exceeds a predetermined link capacity.
15. The non-transitory computer readable medium of claim 13 , wherein a total cost of the set of sourcerouted probes is minimized.
16. The non-transitory computer readable medium of claim 13 , wherein a maximum cost of each sourcerouted probe in the set of source routed probes is minimized.
17. The non-transitory computer readable medium of claim 13 , wherein a total number of sourcerouted probes in the set of sourcerouted probes is minimized.
18. The non-transitory computer readable medium of claim 13 , wherein the set of sourcerouted probes is deployed to monitor a voice over Internet protocol (VoIP) service.