Managing resources using a graph inference model to infer tags
Methods and systems for managing operation of data processing systems are disclosed. A management system may obtain resource data based on operation of the data processing systems while using resources managed by the management system. The resource data may be semantically enhanced based on ontology definitions and represented as a graph structure. The graph structure may be used by a graph inference model (e.g., a graph neural network) to obtain predicted tags for the resources. The graph inference model may be based on a plurality of graph structures that may provide information regarding generalized relationships between at least the resources. The predicted tags may be used by the management system to update operation of the data processing systems.
1. A method of managing operation of data processing systems, the method comprising:
obtaining, by a management system and from the data processing systems, resource data based on operation of the data processing systems while using resources managed by the management system;
identifying, by the management system, enhanced information regarding the resource data based at least in part on ontology definitions, the ontology definitions providing a predefined schema for classifying portions of the resource data;
obtaining, by the management system, at least one graph structure based on the enhanced information, the at least one graph structure comprising nodes based on the resources, and edges based on relationships between the resources;
obtaining, by the management system and using a graph inference model, at least one predicted tag for at least a portion of the graph structure;
servicing, by the management system and using the at least one predicted tag and the at least one graph structure, a request for updating operation of the data processing systems to obtain updated data processing systems; and
providing computer-implemented services using the updated data processing systems.
2. The method of claim 1 , wherein identifying the enhanced information comprises:
mapping each resource of the resources indicated by the resource data according to the predefined schema; and
adding the enhanced information for the each resource based on the ontology definitions.
3. The method of claim 1 , wherein the graph inference model is a graph neural network.
4. The method of claim 3 , wherein the graph neural network is based on a plurality of graph structures that are semantically enhanced with validated tags for at least a portion of the plurality of graph structures.
5. The method of claim 4 , wherein the validated tags provide information regarding generalized relationships between the at least a portion of the plurality of graph structures.
6. The method of claim 4 , wherein obtaining the predicted tag comprises:
inferring, using the graph neural network, the predicted tag for the at least a portion of the graph structure based on the plurality of graph structures that are semantically enhanced.
7. The method of claim 1 , wherein further comprising:
updating, by the management system, the ontology definitions based on new semantic information identified by an entity tasked with managing the ontology definitions.
8. The method of claim 1 , wherein servicing the request for updating operation of the data processing systems comprises:
identifying a portion of the data processing systems based on one or more tags applied to the portion of the data processing systems;
identifying an action set to perform on the identified portion of the data processing systems based on the request; and
performing the action set to modify operation of the portion of the data processing systems.
9. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing operation of data processing systems, the operations comprising:
obtaining, by a management system and from the data processing systems, resource data based on operation of the data processing systems while using resources managed by the management system;
identifying, by the management system, enhanced information regarding the resource data based at least in part on ontology definitions, the ontology definitions providing a predefined schema for classifying portions of the resource data;
obtaining, by the management system, at least one graph structure based on the enhanced information, the at least one graph structure comprising nodes based on the resources, and edges based on relationships between the resources;
obtaining, by the management system and using a graph inference model, at least one predicted tag for at least a portion of the graph structure;
servicing, by the management system and using the at least one predicted tag and the at least one graph structure, a request for updating operation of the data processing systems to obtain updated data processing systems; and
providing computer-implemented services using the updated data processing systems.
10. The non-transitory machine-readable medium of claim 9 , wherein identifying the enhanced information comprises:
mapping each resource of the resources indicated by the resource data according to the predefined schema; and
adding the enhanced information for the each resource based on the ontology definitions.
11. The non-transitory machine-readable medium of claim 9 , wherein the graph inference model is a graph neural network.
12. The non-transitory machine-readable medium of claim 11 , wherein the graph neural network is based on a plurality of graph structures that are semantically enhanced with validated tags for at least a portion of the plurality of graph structures.
13. The non-transitory machine-readable medium of claim 12 , wherein the validated tags provide information regarding generalized relationships between the at least a portion of the plurality of graph structures.
14. The non-transitory machine-readable medium of claim 12 , wherein obtaining the predicted tag comprises:
inferring, using the graph neural network, the predicted tag for the at least a portion of the graph structure based on the plurality of graph structures that are semantically enhanced.
15. A data processing system, comprising:
a processor; and
a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing operation of data processing systems, the operations comprising:
obtaining, by a management system and from the data processing systems, resource data based on operation of the data processing systems while using resources managed by the management system;
identifying, by the management system, enhanced information regarding the resource data based at least in part on ontology definitions, the ontology definitions providing a predefined schema for classifying portions of the resource data;
obtaining, by the management system, at least one graph structure based on the enhanced information, the at least one graph structure comprising nodes based on the resources, and edges based on relationships between the resources;
obtaining, by the management system and using a graph inference model, at least one predicted tag for at least a portion of the graph structure;
servicing, by the management system and using the at least one predicted tag and the at least one graph structure, a request for updating operation of the data processing systems to obtain updated data processing systems; and
providing computer-implemented services using the updated data processing systems.
16. The data processing system of claim 15 , wherein identifying the enhanced information comprises:
mapping each resource of the resources indicated by the resource data according to the predefined schema; and
adding the enhanced information for the each resource based on the ontology definitions.
17. The data processing system of claim 15 , wherein the graph inference model is a graph neural network.
18. The data processing system of claim 17 , wherein the graph neural network is based on a plurality of graph structures that are semantically enhanced with validated tags for at least a portion of the plurality of graph structures.
19. The data processing system of claim 18 , wherein the validated tags provide information regarding generalized relationships between the at least a portion of the plurality of graph structures.
20. The data processing system of claim 18 , wherein obtaining the predicted tag comprises:
inferring, using the graph neural network, the predicted tag for the at least a portion of the graph structure based on the plurality of graph structures that are semantically enhanced.