Enriched self-healing for cloud platforms

Patent metadata

Field	Value
Publication number	US-10601640-B1
Application number	US-201916421165-A
Country	US
Kind code	B1
Filing date	May 23, 2019
Priority date	May 23, 2019
Publication date	Mar 24, 2020
Grant date	Mar 24, 2020

Abstract

An applied intelligence framework may receive log information descriptive of a cloud computing stack. The applied intelligence framework may generate a stack token. The stack token may include a computer resource node representative a computer resource of the cloud computing stack. The applied intelligence framework may access, from an ontology repository, a diagnosis instruction. The diagnosis instruction may determine a fault based on at least one of the log parameters. The applied intelligence framework may execute the diagnosis instruction to determine the fault. The applied intelligence framework may append, to the stack token, a fault node representative of the fault. The applied intelligence framework may query an ontology repository based on the stack token to identify a resolution identifier. The applied intelligence framework may append, to the stack token, a resolution node. The applied intelligence framework may determine, based on stack token and the applied ontology repository, a resolution to the fault.

First claim

The invention claimed is:

1. A method, comprising
receiving, by a cloud computing environment in communication with a cloud computing stack, log information descriptive of one or more computer resources of the cloud computing stack;
extracting, from the log information, log parameters, the log parameters comprising a computer resource identifier corresponding to a computer resource and information indicative of operation of the computer resource;
generating a stack token, the stack token comprising the computer resource identifier representative of the computer resource;
accessing, from an ontology repository, a diagnosis instruction, the diagnosis instruction configured to determine presence of a fault on the cloud computing stack based on at least one of the log parameters;
executing the diagnosis instruction to determine the presence of the fault;
appending, to the stack token, a fault identifier representative of the fault;
querying the ontology repository based on the stack token to identify a resolution identifier, the resolution identifier representative of a resolution to the fault;
appending, to the stack token, the resolution identifier;
determining, based on the stack token and a confidence metric, a resolution instruction executable by the cloud computing stack to communicate with the computer resource identified in the stack token and respond to the fault after the fault has occurred;
communicating the resolution instruction to the cloud computing stack to cause the cloud computing stack to execute the resolution instruction;
receiving, from the cloud computing stack, a fault resolution result generated by the cloud computing stack, the fault resolution result indicative of the fault being successfully resolved or unsuccessfully resolved in response to execution of the resolution instruction by the cloud computing stack; and
calibrating the confidence metric based on the fault resolution result.

2. The method of claim 1 , further comprising:
appending, to the stack token, the fault resolution result.

3. The method of claim 2 , wherein the ontology repository comprises a knowledge graph, the knowledge graph comprising a plurality of nodes and edges, the method further comprising:
enriching the knowledge graph by appending, to the knowledge graph, information from the stack token, the information from the stack token comprising the fault identifier, the resolution identifier, the computer resource identifier, the fault resolution result or a combination thereof.

4. The method of claim 1 , wherein accessing, from the ontology repository, the diagnosis instruction further comprises:
selecting the diagnosis instruction in response to the diagnosis instruction being mapped to at least one of the log parameters in the ontology repository.

5. The method of claim 1 , wherein querying the ontology repository based on the stack token to identify the resolution identifier further comprises:
extracting, from the stack token, the computer resource identifier, a fault identifier indicative of the fault, or a combination thereof;
generating a query request comprising the computer resource identifier, the fault identifier, or any combination thereof;
submitting the query request to the ontology repository; and
identifying, in the ontology repository, the resolution identifier based on the query request.

6. The method of claim 1 , wherein determining, based on the stack token and the confidence metric, the resolution instruction executable by the cloud computing stack to communicate with the computer resource identified in the stack token and respond to the fault further comprises:
extracting, from the stack token, the resolution identifier and the computer resource identifier;
identifying, in the ontology repository, a plurality of resolution instructions mapped with the resolution identifier, each of the resolution instruction respectively configured to communicate with different corresponding computer resources to respond to the fault; and
selecting, from the resolution instructions, the resolution instruction based on the computer resource identifier of the stack token.

7. The method of claim 1 , wherein the stack token comprises a graph data structure, wherein the graph data structure comprises a computer resource node, a fault node, and a resolution node, wherein the computer resource node comprises the computer resource identifier, the fault node comprises the fault identifier, and the resolution node comprises the resolution identifier.

8. A system comprising:
a cloud computing environment in communication with a plurality of cloud computing stacks, the cloud computing environment comprising a processor, the processor configured to:
receive log information descriptive of one or more computer resources of a cloud computing stack included among the cloud computing stacks;
extract, from the log information, log parameters, the log parameters comprising a computer resource identifier corresponding to a computer resource of the cloud computing stack and information indicative of operation of the computer resource;
generate a stack token, the stack token comprising a computer resource identifier representative of the computer resource;
access, from an ontology repository, a diagnosis instruction, the diagnosis instruction configured to determine an occurrence of a fault by the cloud computing stack based on at least one of the log parameters;
execute the diagnosis instruction to determine the occurrence of the fault;
append, to the stack token, a fault identifier representative of the fault;
query an ontology repository based on the stack token to identify a resolution identifier, the resolution identifier representative of a resolution to the fault;
append, to the stack token, the resolution identifier;
determine a resolution instruction based on the stack token and a confidence metric, the resolution instruction being executable by the cloud computing stack to communicate with the computer resource to respond to the fault after the occurrence of the fault;
communicate the resolution instruction to the cloud computing stack to cause the cloud computing stack to execute the resolution instruction and resolve the fault;
receive, from the cloud computing stack, a fault resolution result generated by the cloud computing stack, the fault resolution result indicative of the fault being successfully resolved or unsuccessfully resolved in response to execution of the resolution instruction by the cloud computing stack; and
adjust the confidence metric based on the fault resolution result.

9. The system of claim 8 , wherein the cloud computing environment is physically remote from the plurality of cloud computing stacks.

10. The system of claim 8 , wherein the ontology repository comprises information from a plurality of stack tokens respectively generated by different corresponding cloud computing stacks.

11. The system of claim 8 , wherein the processor is further configured to:
append, to the stack token, the fault resolution result.

12. The system of claim 11 , wherein the ontology repository comprises a knowledge graph, the knowledge graph comprising a plurality of nodes and edges, wherein the processor is further configured to:
append, to the knowledge graph, information from the stack token, the information from the stack token comprising the fault identifier, the resolution identifier, the computer resource identifier, the fault resolution result, or a combination thereof.

13. The system of claim 8 , wherein to access, from the ontology repository, the diagnosis instruction, the processor is further configured

Assignees

• Accenture Global Solutions Ltd

Inventors

• Das Jagaran
• Khan Rizwan
• Radhakrishnan Biju

Classifications

• H04L41/22

  comprising specially adapted graphical user interfaces [GUI] · CPC title

• H04L41/145

  involving simulating, designing, planning or modelling of a network · CPC title

• H04L41/069 Primary

  using logs of notifications; Post-processing of notifications · CPC title

• H04L41/0631

  using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis · CPC title

• H04L41/0636

  based on a decision tree analysis · CPC title