Methods, systems and computer readable media for diagnosing network function virtualization performance
Performance issues in a service function chain having a plurality of resources and a plurality of network functions each having a network function queue are diagnosed. Each network function queue is monitored and queueing information for input packets for each of the plurality of network functions is dumped to a data store. Each resource that is under contention is identified as well as which of the network functions is a contender for the resources. A diagnosing algorithm is used to diagnose performance problems and an impact graph for each victim packet is generated. A summary of results as a list of rules is then provided.
1. A method for diagnosing performance issues in a service function chain having a plurality of resources and wherein the service function chain has a plurality of network functions, each of the plurality of network functions having a network function queue, comprising:
monitoring each network function queue;
dumping queueing information for input packets for each of the plurality of network functions;
identifying, from the dumped queueing information, each of the plurality of resources that is under contention;
determining which of the plurality of network functions is a contender for the each of the plurality of resources;
using a diagnosing algorithm to diagnose performance problems based on the determining step;
generating an impact graph for each victim packet identified by the algorithm as a result of a performance problem; and
generating a summary of results of the diagnosing algorithm as a list of rules.
2. The method of claim 1 wherein the performance issues comprise long tail latency or packet drops.
3. The method of claim 1 wherein the step of using a diagnosing algorithm comprises providing an impact score for each of the plurality of resources that is under contention.
4. The method of claim 1 wherein the step of identifying each of the plurality of resources that is under contention comprises using resource counters to determine how each of the plurality of resources is under contention.
5. The method of claim 4 wherein the resource counters comprise symptom counters, resource measurement counters and traffic measurement counters.
6. The method of claim 1 wherein the step of using a diagnosing algorithm comprises running a score back-propagation step for determining which of the plurality of network functions propagate an impact to a victim packet.
7. The method of claim 6 further comprising running the score back-propagation step recursively until all root nodes are back-propagated.
8. A system for diagnosing performance issues in a service function chain comprising:
a plurality of resources;
a plurality of network functions each having a network function queue;
a processor;
a non-volatile computer memory for storing computer instruction coupled to the processor, wherein processor, responsive to executing the computer instructions, performs operations comprising:
monitoring each network function queue;
dumping queueing information for input packets for each of the plurality of network functions;
identifying, from the dumped queueing information, each of the plurality of resources that is under contention;
determining which of the plurality of network functions is a contender for the each of the plurality of resources;
using a diagnosing algorithm to diagnose performance problems based on the determining step;
generating an impact graph for each victim packet identified by the algorithm as a result of a performance problem; and
generating a summary of results of the diagnosing algorithm as a list of rules.
9. The system of claim 8 wherein the performance issues comprise long tail latency or packet drops.
10. The system of claim 8 wherein using a diagnosing algorithm comprises providing an impact score for each of the plurality of resources that is under contention.
11. The system of claim 8 wherein identifying each of the plurality of resources that is under contention comprises using resource counters to determine how each of the plurality of resources is under contention.
12. The system of claim 11 wherein the resource counters comprise symptom counters, resource measurement counters and traffic measurement counters.
13. The system of claim 8 wherein using a diagnosing algorithm comprises running a score back-propagation step for determining which of the plurality of network functions propagate an impact to a victim packet.
14. The system of claim 13 wherein the operations further comprise running the score back-propagation step recursively until all root nodes are back-propagated.
15. A non-transitory, tangible computer-readable medium having computer-executable instructions stored thereon which, when executed by a computer, cause the computer to perform a method for diagnosing performance issues in a service function chain having a plurality of resources and wherein the service chain has a plurality of network functions, each of the plurality of network functions having a network function queue, comprising:
monitoring each network function queue;
dumping queueing information for input packets for each of the plurality of network functions;
identifying, from the dumped queueing information, each of the plurality of resources that is under contention;
determining which of the plurality of network functions is a contender for the each of the plurality of resources;
using a diagnosing algorithm to diagnose performance problems based on the determining step;
generating an impact graph for each victim packet identified by the algorithm as a result of a performance problem; and
generating a summary of results of the diagnosing algorithm as a list of rules.
16. The non-transitory, tangible computer-readable medium of claim 15 wherein the performance issues comprise long tail latency or packet drops.
17. The non-transitory, tangible computer-readable medium of claim 15 wherein using a diagnosing algorithm comprises providing an impact score for each of the plurality of resources that is under contention.
18. The non-transitory, tangible computer-readable medium of claim 15 wherein identifying each of the plurality of resources that is under contention comprises using resource counters to determine how each of the plurality of resources is under contention.
19. The non-transitory, tangible computer-readable medium of claim 18 wherein the resource counters comprise symptom counters, resource measurement counters and traffic measurement counters.
20. The non-transitory, tangible computer-readable medium of claim 15 wherein using a diagnosing algorithm comprises running a score back-propagation step for determining which of the plurality of network functions propagate an impact to a victim package.