Method and apparatus for determining causality mapping of distributed systems
View Patent ↗A method and apparatus for determining causality mapping between causing events and detectable events among a plurality of nodes in a distributed system is disclosed. The method comprises the steps of automatically generating a causality mapping model of the dependences between causing events at the nodes of the distributed system and the detectable events in a subset of the nodes, the model suitable for representing the execution of at least one system operation. In one aspect the generation is perform by selecting nodes associated with each of the detectable events from the subset of the nodes and indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node is a known distance from at least one node selected from the selected nodes. In still another aspect, the processing described herein is in the form of a computer-readable medium suitable for providing instruction to a computer or processing system for executing the processing claimed.
1. A computer implemented method for determining causality mapping between causing events and detectable events among a plurality of nodes in a distributed system comprising physical devices, the method comprising the steps of:
automatically generating, in one or more computers, a causality mapping model, the mapping model mapping the dependences between causing events at the nodes and detectable events associated with a subset of the nodes, the model suitable for representing the execution of at least one system operation; wherein each causing event of the causing events corresponds to a cause of a problem in the distributed system; wherein each detectible event of the detectible events corresponds to a symptom of at least one problem in the distributed system; and wherein the mapping model may be used in determining at least one causing event contributing to the cause of the detected event;
wherein the step of generating the model comprises the steps of:
selecting nodes associated with each of the detectable events from the subset of the nodes;
indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node divides the system into a plurality of disjoint partitions and at least one of the nodes associated with a selected detectable event is contained in one of the disjoint partitions and at least one other of the nodes associated with the selected detectable event is contained in another one of the disjoint partitions.
2. The method as recited in claim 1 , wherein the step of generating the model comprises the steps of: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between causing events and each of the detectable events for each causing event at a node when the causing event node is in one of the routes between the selected nodes.
3. The method as recited in claim 1 , wherein the step of generating the model comprises the step of:
indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node is a known distance from at least one node selected from the selected nodes.
4. The method as recited in claim 1 , wherein the step of generating the model comprises the steps of: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node divides the system into a plurality of disjoint partitions and at least one of the nodes associated with a selected detectable event is contained in one of the disjoint partitions and at least one other of the nodes associated with the selected detectable event is contained in another one of the disjoint partitions.
5. The method as recited in claim 1 , wherein the step of generating the model comprises the steps of: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between a causing event and at least one detectable event for each causing event when the causing event node is a known distance from at least one node associated with the detectable event, wherein the dependency is a function of the number of times the causing event node appears in the routes between the nodes.
6. The method as recited in claim 1 , wherein the system operation is selected from the group consisting of: fault detection, fault monitoring, performance, congestion, connectivity, interface failure, node failure, link failure, routing protocol error, routing control errors, and root-cause analysis.
7. The method as recited in claim 1 , wherein the dependency represents a probability a detected event was caused by a causing event.
8. The method as recited in claim 7 , wherein the probability decreases as a function of the number nodes from the causing event node.
9. The method as recited in claim 1 , wherein the dependency is a predetermined value.
10. The method as recited in claim 1 , wherein model is selected from the group consisting of: a causality matrix and causality graph.
11. The method as recited in claim 1 , wherein the path between nodes selected from the subset of nodes represents the shortest paths.
12. The method as recited in claim 1 , wherein the dependency is a measure of the number of times a causing event node appears in the total number of paths between the nodes selected from the subset of nodes.
13. An apparatus for determining causality mapping between causing events and detectable events among a plurality of nodes in a distributed system, the apparatus comprising: a processor in communication with a memory, the processor executing code for: automatically generating a causality mapping model, the mapping model mapping the dependences between causing events at the nodes and detectable events associated with a subset of the nodes, the model suitable for representing the execution of at least one system operation; wherein each causing event of the causing events corresponds to a cause of a problem in the distributed system; wherein each detectible event of the detectible events corresponds to a symptom of at least one problem in the distributed system; and wherein the mapping model may be used in determining at least one causing event contributing to the cause of the detected event; wherein the processor further executing code for generating the model comprising:
selecting nodes associated with each of the detectable events from the subset of the nodes;
indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node divides the system into a plurality of disjoint partitions and at least one of the nodes associated with a selected detectable event is contained in one of the disjoint partitions and at least one other of the nodes associated with the selected detectable event is contained in another one of the disjoint partitions.
14. The apparatus as recited in claim 13 , the processor further executing code for generating the model comprising: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between causing events and each of the detectable events for each causing event at a node when the causing event node is in one of the routes between the selected nodes.
15. The apparatus as recited in claim 13 , the processor further executing code for generating the model comprising:
indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node is a known distance from at least one node selected from the selected nodes.
16. The apparatus as recited in claim 13 , wherein the processor further executing code for generating the model comprising: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node divides the system into a plurality of disjoint partitions and at least one of the nodes associated with a selected detectable event is contained in one of the disjoint partitions and at least one other of the nodes associated with the selected detectable event is contained in another one of the disjoint partitions.
17. The apparatus as recited in claim 13 , the processor further executing code for generating the model comprising: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between a causing event and at least one detectable event for each causing event when the causing event node is a known distance from at least one node associated with the detectable event, wherein the dependency is a function of the number of times the causing event node appears in the routes between the nodes.
18. The apparatus as recited in claim 13 , wherein the system operation is selected from the group consisting of: fault detection, fault monitoring, interface failure, node failure, link failure, routing protocol error, routing control errors, and root-cause analysis.
19. The apparatus as recited in claim 13 , wherein the dependency represents a probability a detected event was caused by a causing event.
20. The apparatus as recited in claim 13 , wherein the indication is a predetermined value.
21. The apparatus as recited in claim 13 , wherein model is selected from the group consisting of: a causality matrix and causality graph.
22. The apparatus as recited in claim 13 , wherein the path between nodes selected from the subset of nodes represents the shortest paths.
23. The apparatus as recited in claim 13 , wherein the dependency is a measure of the number of times a causing event node appears in the total number of paths between the nodes selected from the subset of nodes.
24. A non-transitory computer-readable medium containing code for determining causality mapping between causing events and detectable events among a plurality of nodes in a distributed system, the code providing instruction to a processing system for executing code for: automatically generating a causality mapping model, the mapping model mapping the dependences between causing events at the nodes and detectable events associated with a subset of the nodes, the model suitable for representing the execution of at least one system operation; wherein each causing event of the causing events corresponds to a cause of a problem in the distributed system; wherein each detectible event of the detectible events corresponds to a symptom of at least one problem in the distributed system; and wherein the mapping model may be used in determining at least one causing event contributing to the cause of the detected event.
25. The non-transitory computer-readable medium as recited in claim 24 , the code providing instruction to the processing system for generating the model comprising: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between causing events and each of the detectable events for each causing event at a node when the causing event node is in one of the routes between the selected nodes.
26. The non-transitory computer-readable medium as recited in claim 24 the code providing instruction to the processing system for generating the model comprising:
indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node is a known distance from at least one node selected from the selected nodes.
27. The non-transitory computer-readable medium as recited in claim 24 , the code providing instruction to the processing system for generating the model comprising: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between a causing event and at least one detectable event for each causing event at a node when the causing event node divides the system into a plurality of disjoint partitions and at least one of the nodes associated with a selected detectable event is contained in one of the disjoint partitions and at least one other of the nodes associated with the selected detectable event is contained in another one of the disjoint partitions.
28. The non-transitory computer-readable medium as recited in claim 24 , the code providing instruction to the processing system for generating the model comprising: selecting nodes associated with each of the detectable events from the subset of the nodes; and indicating the dependency between a causing event and at least one detectable event when the causing event node is a known distance from at least one node associated with the detectable event, wherein the dependency is a function of the number of times the causing event node appears in the routes between the nodes.
29. The non-transitory computer-readable medium as recited in claim 24 , wherein the system operation is selected from the group consisting of: fault detection, fault monitoring, failure, node failure, link failure, routing protocol error, routing control errors, and root-cause analysis.
30. The non-transitory computer-readable medium as recited in claim 24 , wherein the dependency represents a probability a detected event was caused by a causing event.
31. The non-transitory computer-readable medium as recited in claim 24 , wherein the probability decreases as a function of the number nodes from the causing event node.
32. The non-transitory computer-readable medium as recited in claim 24 , wherein the dependency is a predetermined value.
33. The non-transitory computer-readable medium as recited in claim 24 , wherein model is selected from the group consisting of: a causality matrix and causality graph.
34. The non-transitory computer-readable medium as recited in claim 24 , wherein the path between nodes selected from the subset of nodes represents the shortest paths.
35. The non-transitory computer-readable medium as recited in claim 24 , wherein the dependency is a measure of the number of times a causing event node appears in the total number of paths between the nodes selected from the subset of nodes.