System for inferring dependencies among computing systems
Systems and methods are described for identifying, tracking, and customizing dependencies between components of a computing environment. By providing greater insight and transparency into dependencies, the systems and methods can facilitate modeling the underlying architecture of applications and computer hardware. As a result, IT personnel can better track relationships between components. Custom filtering tools can also provide IT personnel with tools to switch from different types of dependency views that focus on application-oriented views, hardware-oriented views, or other custom views. Model annotation tools can also enable IT personnel to customize a dependency model to reflect real-world application and hardware monitoring conditions.
1. A method of providing information about dependencies in a computing environment, the method comprising, by a computer system comprising computer hardware:
receiving monitoring data obtained from a plurality of monitored resources in a computing environment;
transforming the monitoring data into a topology model comprising a plurality of interconnected topology objects, the plurality of interconnected topology objects representing the plurality of monitored resources, wherein interconnections between the plurality of interconnected topology objects reflect existing relationships between the plurality of monitored resources in the computing environment; and
storing the topology model in a data store within the computing environment;
detecting a first dependency between first and second processes corresponding to first and second topology objects of the interconnected topology objects based, at least in part, on an observed communication between the first and second processes, the observed communication between the first and second processes being at least partially different from the existing relationships represented in the stored topology model, and wherein the first process is running on a first virtual machine and the second process is running on a second virtual machine;
responsive to the detecting the first dependency, looking up the first and second processes in the stored topology model to determine related model objects, the looking up comprising following at least one of ancestor and descendant links of the first and second topology objects during a traversal of the stored topology model;
responsive to the looking up, identifying a third topology object that is at least one of an ancestor and a descendant of the first topology object and a fourth topology object that is at least one of an ancestor and a descendant of the second topology object, wherein the third topology object corresponds to the first virtual machine on which the first process is running and the fourth topology object corresponds to the second virtual machine on which the second process is running;
inferring, based at least in part on the observed communication and the traversal of the stored topology model, a second dependency between the first and second virtual machines corresponding to the third and fourth topology objects of the interconnected topology objects;
storing data representing the first and second dependencies in a dependency graph; and
outputting a user interface comprising at least a portion of the dependency graph.
2. The method of claim 1 , comprising ignoring selected topology objects of the interconnected topology objects during the traversal of the stored topology model when inferring the second dependency in response to detecting a registry setting specifying that the selected topology objects are to be ignored.
3. The method of claim 1 , wherein the outputting comprises outputting a registry user interface configured to enable users to annotate the dependency graph to adjust one or both of the first and second dependencies represented in the dependency graph to thereby produce a modified dependency graph.
4. The method of claim 3 , comprising:
receiving via the registry user interface a time-sensitive registry setting, the time-sensitive registry setting indicating a scheduled period during which selected ones of the interconnected topology objects are scheduled to be offline;
inferring that any child or descendent objects from the selected ones of the interconnected topology objects are also offline during the scheduled period; and
outputting a portion of the modified dependency graph for presentation to a user, wherein the selected ones of the interconnected topology objects and the inferred child or descendent objects are not provided in the outputted portion of the modified dependency graph during the scheduled period due to the time-sensitive registry setting.
5. The method of claim 4 , comprising outputting the portion of the dependencies in a graph or a table.
6. The method of claim 3 , wherein the dependency graph comprises a tree data structure configured to link the interconnected topology objects in ancestor-descendent relationships, and wherein the registry user interface is further configured to provide functionality that enables users to adjust the ancestor-descendent relationships.
7. The method of claim 3 , comprising enabling a user to select a subset of the modified dependency graph for presentation.
8. The method of claim 3 , comprising overriding at least one registry variable based on a custom filter to further modify a view of the dependency graph.
9. The method of claim 3 , wherein the dependency graph comprises selected interconnected topology objects of the interconnected topology objects arranged in a hierarchical data structure.
10. The method of claim 9 , wherein the registry user interface is further configured to enable users to modify relationships between the selected interconnected topology objects in the hierarchical data structure.
11. A system comprising a processor and memory, wherein the processor and memory in combination are operable to implement a method comprising:
receiving monitoring data obtained from a plurality of monitored resources in a computing environment;
transforming the monitoring data into a topology model comprising a plurality of interconnected topology objects, the plurality of interconnected topology objects representing the plurality of monitored resources, wherein interconnections between the plurality of interconnected topology objects reflect existing relationships between the plurality of monitored resources in the computing environment; and
storing the topology model in a data store within the computing environment;
detecting a first dependency between first and second processes corresponding to first and second topology objects of the interconnected topology objects based, at least in part, on an observed communication between the first and second processes, the observed communication between the first and second processes being at least partially different from the existing relationships represented in the stored topology model, and wherein the first process is running on a first virtual machine and the second process is running on a second virtual machine;
responsive to the detecting the first dependency, looking up the first and second processes in the stored topology model to determine related model objects, the looking up comprising following at least one of ancestor and descendant links of the first and second topology objects during a traversal of the stored topology model;
responsive to the looking up, identifying a third topology object that is at least one of an ancestor and a descendant of the first topology object and a fourth topology object that is at least one of an ancestor and a descendant of the second topology object, wherein the third topology object corresponds to the first virtual machine on which the first process is running and the fourth topology object corresponds to the second virtual machine on which the second process is running;
inferring, based at least in part on the observed communication and the traversal of the stored topology model, a second dependency between the first and second virtual machines corresponding to the third and fourth topology objects of the interconnected topology objects;
storing data representing the first and second dependencies in a dependency graph; and
outputting a user interface comprising at least a portion of the dependency graph.
12. The system of claim 11 , the method comprising ignoring selected topology objects of the interconnected topology objects during the traversal of the stored topology model when inferring the second dependency in response to detecting a registry setting specifying that the selected topology objects are to be ignored.
13. The system of claim 11 , wherein the outputting comprises outputting a registry user interface configured to enable users to annotate the dependency graph to adjust one or both of the first and second dependencies represented in the dependency graph to thereby produce a modified dependency graph.
14. The system of claim 13 , the method comprising:
receiving via the registry user interface a time-sensitive registry setting, the time-sensitive registry setting indicating a scheduled period during which selected ones of the interconnected topology objects are scheduled to be offline;
inferring that any child or descendent objects from the selected ones of the interconnected topology objects are also offline during the scheduled period; and
outputting a portion of the modified dependency graph for presentation to a user, wherein the selected ones of the interconnected topology objects and the inferred child or descendent objects are not provided in the outputted portion of the modified dependency graph during the scheduled period due to the time-sensitive registry setting.
15. The system of claim 14 , the method comprising outputting the portion of the dependencies in a graph or a table.
16. The system of claim 13 , wherein the dependency graph comprises a tree data structure configured to link the interconnected topology objects in ancestor-descendent relationships, and wherein the registry user interface is further configured to provide functionality that enables users to adjust the ancestor-descendent relationships.
17. The system of claim 13 , the method comprising enabling a user to select a subset of the modified dependency graph for presentation.
18. The system of claim 13 , the method comprising overriding at least one registry variable based on a custom filter to further modify a view of the dependency graph.
19. The system of claim 13 , wherein the dependency graph comprises selected interconnected topology objects of the interconnected topology objects arranged in a hierarchical data structure.
20. A computer-program product comprising a non-transitory computer-usable medium having computer-readable program code embodied therein, the computer-readable program code adapted to be executed to implement a method on a particular recording device in a peer-to-peer recording system comprising a plurality of networked recording devices, the method comprising:
receiving monitoring data obtained from a plurality of monitored resources in a computing environment;
transforming the monitoring data into a topology model comprising a plurality of interconnected topology objects, the plurality of interconnected topology objects representing the plurality of monitored resources, wherein interconnections between the plurality of interconnected topology objects reflect existing relationships between the plurality of monitored resources in the computing environment; and
storing the topology model in a data store within the computing environment;
detecting a first dependency between first and second processes corresponding to first and second topology objects of the interconnected topology objects based, at least in part, on an observed communication between the first and second processes, the observed communication between the first and second processes being at least partially different from the existing relationships represented in the stored topology model, and wherein the first process is running on a first virtual machine and the second process is running on a second virtual machine;
responsive to the detecting the first dependency, looking up the first and second processes in the stored topology model to determine related model objects, the looking up comprising following at least one of ancestor and descendant links of the first and second topology objects during a traversal of the stored topology model;
responsive to the looking up, identifying a third topology object that is at least one of an ancestor and a descendant of the first topology object and a fourth topology object that is at least one of an ancestor and a descendant of the second topology object, wherein the third topology object corresponds to the first virtual machine on which the first process is running and the fourth topology object corresponds to the second virtual machine on which the second process is running;
inferring, based at least in part on the observed communication and the traversal of the stored topology model, a second dependency between the first and second virtual machines corresponding to the third and fourth topology objects of the interconnected topology objects;
storing data representing the first and second dependencies in a dependency graph; and
outputting a user interface comprising at least a portion of the dependency graph.