Cloud infrastructure planning assistant via multi-agent AI

Patent metadata

Field	Value
Publication number	US-11516091-B2
Application number	US-201916390089-A
Country	US
Kind code	B2
Filing date	Apr 22, 2019
Priority date	Apr 22, 2019
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

Abstract

Cloud infrastructure planning systems and methods can utilize artificial intelligence/machine learning agents for developing a plan of demand, plan of record, plan of execution, and plan of availability for developing cloud infrastructure plans that are more precise and accurate, and that learn from previous planning and deployments. Some agents include one or more of supervised, unsupervised, and reinforcement machine learning to develop accurate predictions and perform self-tuning alone or in conjunction with other agents.

First claim

What is claimed is:

1. A cloud infrastructure planning system, comprising:
a plan of demand assistant configured to generate a site solution to a forecasted capacity demand set, wherein the site solution is based on a capacity correlation derived from a historical site solution data set;
a plan of record advisor configured to generate a plan of record for the site solution, wherein the plan of record is based on an infrastructure correlation derived from a historical infrastructure data set;
a plan of execution analyzer configured to generate an execution design defining equipment meeting the plan of record for a planned buildout, wherein the execution design is based on an equipment correlation derived from a historical equipment data set;
a plan of availability evaluator configured to generate a resource prediction defining a service level based on the execution design, wherein the resource prediction is based on an availability correlation derived from a historical availability data set.

2. The cloud infrastructure planning system of claim 1 , wherein the capacity correlation is determined by supervised machine learning.

3. The cloud infrastructure planning system of claim 1 , wherein the infrastructure correlation is determined by unsupervised machine learning.

4. The cloud infrastructure planning system of claim 1 , wherein the equipment correlation is determined by one or both of supervised and unsupervised machine learning.

5. The cloud infrastructure planning system of claim 1 , wherein the availability correlation is determined by supervised machine learning.

6. The cloud infrastructure planning system of claim 1 , further comprising:
a first interface associated with one of the plan of demand assistant, the plan of record advisor, the plan of execution analyzer, and the plan of availability evaluator; and
a second interface associated with another of the plan of demand assistant, the plan of record advisor, the plan of execution analyzer, and the plan of availability evaluator,
wherein the first interface is provided to a first user and the second interface is provided to a second user different from the first user.

7. The cloud infrastructure planning system of claim 1 ,
wherein a first one of the plan of demand assistant, the plan of record advisor, the plan of execution analyzer, and the plan of availability evaluator is effectuated on a first computer system, and
wherein a second one of the plan of demand assistant, the plan of record advisor, the plan of execution analyzer, and the plan of availability evaluator is effectuated on a second computer system different from the first computer system.

8. The cloud infrastructure planning system of claim 1 , wherein the historical site solution data set includes acceptance data concerning a prior site solution generated by the plan of demand assistant, and wherein the acceptance data includes whether a site solution was approved by a user, whether a site solution advanced to subsequent planning, whether a site solution was actually implemented through deployment of a plan based thereon, or a combination thereof.

9. The cloud infrastructure planning system of claim 8 , wherein the historical infrastructure data set includes acceptance data concerning a prior plan of record generated by the plan of record advisor.

10. The cloud infrastructure planning system of claim 9 , wherein the historical equipment data set includes acceptance data concerning a prior execution design generated by the plan of execution analyzer.

11. The cloud infrastructure planning system of claim 10 , wherein the historical equipment data set includes one or more of overbooking data, service utilization data, and virtual network function sizing.

12. The cloud infrastructure planning system of claim 11 , wherein the historical availability data set includes accuracy data concerning a prior resource prediction generated by the plan of availability evaluator.

13. A method, comprising:
receiving cloud infrastructure planning data;
generating a site solution to a forecasted capacity demand set, wherein the site solution is based on the cloud infrastructure planning data and a capacity correlation derived from a historical site solution data set;
generating a plan of record for the site solution implementing a planned buildout, wherein the plan of record is based on an infrastructure correlation derived from a historical infrastructure data set;
generating an execution design defining equipment meeting the plan of record for the planned buildout, wherein the execution design is based on an equipment correlation derived from a historical equipment data set;
generating a resource prediction defining a service level based on the execution design, wherein the resource prediction is based on an availability correlation derived from a historical availability data set.

14. The method of claim 13 , wherein the capacity correlation is determined by supervised machine learning.

15. The method of claim 13 , wherein the infrastructure correlation is determined by unsupervised machine learning.

16. The method of claim 13 , wherein the equipment correlation is determined by one or both of supervised and unsupervised machine learning.

17. The method of claim 13 , wherein the availability correlation is determined by supervised machine learning.

18. The method of claim 13 , comprising:
providing a first interface configured to display a first one of the site solution, the plan of record, the execution design, and the resource prediction; and
providing a second interface configured to display a second one of the site solution, the plan of record, the execution design, and the resource prediction, wherein the second interface is different from the first interface.

19. The method of claim 13 , wherein the historical equipment data set includes one or more of overbooking data, service utilization data, and virtual network function sizing.

20. A non-transitory computer-readable medium storing instructions that when executed by a processor effectuate operations comprising:
receiving cloud infrastructure planning data;
generating a site solution to a forecasted capacity demand set, wherein the site solution is based on the cloud infrastructure planning data and a capacity correlation derived from a historical site solution data set;
generating a plan of record for the site solution implementing a planned buildout, wherein the plan of record is based on an infrastructure correlation derived from a historical infrastructure data set;
generating an execution design defining equipment meeting the plan of record for the planned buildout, wherein the execution design is based on an equipment correlation derived from a historical equipment data set;
generating a resource prediction defining a service level based on the execution design, wherein the resource prediction is based on an availability correlation derived from a historical availability data set.

Assignees

• At & T Ip I Lp

Inventors

• Dome George
• Chen Chuxin
• Oetting John

Classifications

• H04L47/808

  User-type aware · CPC title

• H04L41/5096

  wherein the managed service relates to distributed or central networked applications · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N5/043

  Distributed expert systems; Blackboards · CPC title

• G06N3/006

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title