Granted Patent B2

US 12,298,880 · App. 18/487,024 · Granted May 13, 2025

Inventors: Rajeev Sudhakar Bhopi (Mercer Island, WA); Yiwen Zhu (San Francisco, CA); Helen Mary Serr (Boston, MA); Jonah Karpman (San Diego, CA); Matthew Joseph Gleeson (New York, NY); Nicholas Kent Glaze (Cambridge, MA); Subramaniam Venkatraman Krishnan (San Jose, CA); Irwin Hollar McNeely, III (Cambridge, MA)

Assignee: Microsoft Technology Licensing, LLC.

G06F11/3442G06F8/70

Abstract

Claims (80)

1. A system comprising:

a processor; and

a computer-readable medium storing instructions that are operative upon execution by the processor to:

receive prior-existing utilization data and project metadata, wherein the utilization data comprises capacity information and resource consumption information for prior-existing computational resources, and wherein the project metadata includes information for hierarchically categorizing the prior-existing computational resources;

create, using the utilization data and project metadata, a capacity prediction model for generating a pre-build configuration for a first computational resource;

generate, using the capacity prediction model, the pre-build configuration for the first computational resource; and

tune the pre-build configuration using a selected cost and performance balance point and prior-existing project history data.

2. The system of claim 1 , wherein the instructions are further operative to:

build the first computational resource in accordance with the pre-build configuration, the pre-build configuration comprising a processor count, amount of memory, and/or storage capacity for the first computational resource.

3. The system of claim 2 , wherein the instructions are further operative to:

receive input data by the first computational resource; and

generate output data using the first computational resource, based on at least the input data.

4. The system of claim 1 ,

wherein the capacity information comprises processor count, amount of memory, and/or storage capacity for the prior-existing computational resources;

wherein the resource consumption information comprises slack information and throttling information for the prior-existing computational resources; and

wherein the project history data comprises requested changes or reported incidents for the prior-existing computational resources.

5. The system of claim 1 , wherein the instructions are further operative to:

train a first model to perform capacity sizing for the capacity prediction model based on at least the utilization data;

train a second model to perform workload prediction for the capacity prediction model based on at least the project metadata; and

train a third model to perform capacity sizing for the capacity prediction model based on at least the project history data.

6. The system of claim 1 , wherein the instructions are further operative to:

present a user interface (UI);

receive, through the UI, an initial build target selection, wherein generating the pre-build configuration for the first computational resource comprises generating the pre-build configuration for the first computational resource based on at least the initial build target selection;

receive, through the UI, the selected cost and performance balance point; and

display, in the UI, at least a portion of the pre-build configuration.

7. The system of claim 6 , wherein the instructions are further operative to:

display, in the UI, at least a portion of a hierarchy of the capacity prediction model; and

display, in the UI, information for a prior-existing computational resource used in generating the capacity prediction model.

8. A computer-implemented method comprising:

receiving prior-existing utilization data and project metadata, wherein the utilization data comprises capacity information and resource consumption information for prior-existing computational resources, and wherein the project metadata includes information for hierarchically categorizing the prior-existing computational resources;

creating, using the utilization data and project metadata, a capacity prediction model for generating a pre-build configuration for a first computational resource;

generating, using the capacity prediction model, the pre-build configuration for the first computational resource, the pre-build configuration comprises processor count, amount of memory, and/or storage capacity for the first computational resource; and

tuning the pre-build configuration using a selected cost and performance balance point and prior-existing project history data.

9. The computer-implemented method of claim 8 , further comprising:

building the first computational resource in accordance with the pre-build configuration;

receiving input data by the first computational resource; and

generating output data using the first computational resource, based on at least the input data.

10. The computer-implemented method of claim 8 ,

wherein the capacity information comprises processor count, amount of memory, and/or storage capacity for the prior-existing computational resources;

wherein the resource consumption information comprises slack information and throttling information for the prior-existing computational resources; and

wherein the project history data comprises requested changes or reported incidents for the prior-existing computational resources.

11. The computer-implemented method of claim 8 , further comprising:

training a first model to perform capacity sizing for the capacity prediction model based on at least the utilization data;

training a second model to perform workload prediction for the capacity prediction model based on at least the project metadata; and

training a third model to perform capacity sizing for the capacity prediction model based on at least the project history data.

12. The computer-implemented method of claim 8 , further comprising:

presenting a user interface (UI);

receiving, through the UI, an initial build target selection, wherein generating the pre-build configuration for the first computational resource comprises generating the pre-build configuration for the first computational resource based on at least the initial build target selection;

receiving, through the UI, the selected cost and performance balance point; and

displaying, in the UI, at least a portion of the pre-build configuration.

13. The computer-implemented method of claim 12 , further comprising:

displaying, in the UI, at least a portion of a hierarchy of the capacity prediction model; and

displaying, in the UI, information for a prior-existing computational resource used in generating the capacity prediction model.

14. The computer-implemented method of claim 8 , wherein generating the pre-build configuration and tuning the pre-build configuration each comprises optimizing the pre-build configuration for a target slack rate and a target throttling rate.

15. A computer storage device having computer-executable instructions stored thereon, which, on execution by a computer, cause the computer to perform operations comprising:

receiving prior-existing utilization data and project metadata, wherein the utilization data comprises capacity information and resource consumption information for prior-existing computational resources, and wherein the project metadata includes information for hierarchically categorizing the prior-existing computational resources;

creating, using the utilization data and project metadata, a capacity prediction model for generating a pre-build configuration for a first computational resource;

generating, using the capacity prediction model, the pre-build configuration for the first computational resource;

tuning the pre-build configuration using a selected cost and performance balance point and prior-existing project history data; and

building the first computational resource in accordance with the pre-build configuration, the pre-build configuration comprising a processor count, amount of memory, and/or storage capacity for the first computational resource.

16. The computer storage device of claim 15 , wherein the operations further comprise:

receiving input data by the first computational resource; and

generating output data using the first computational resource, based on at least the input data.

17. The computer storage device of claim 15 , wherein the operations further comprise:

wherein the capacity information comprises processor count, amount of memory, and/or storage capacity for the prior-existing computational resources;

wherein the resource consumption information comprises slack information and throttling information for the prior-existing computational resources; and

wherein the project history data comprises requested changes or reported incidents for the prior-existing computational resources.

18. The computer storage device of claim 15 , wherein the operations further comprise:

training a first model to perform capacity sizing for the capacity prediction model based on at least the utilization data;

training a second model to perform workload prediction for the capacity prediction model based on at least the project metadata; and

training a third model to perform capacity sizing for the capacity prediction model based on at least the project history data.

19. The computer storage device of claim 15 , wherein the operations further comprise:

presenting a user interface (UI);

receiving, through the UI, an initial build target selection, wherein generating the pre-build configuration for the first computational resource comprises generating the pre-build configuration for the first computational resource based on at least the initial build target selection;

receiving, through the UI, the selected cost and performance balance point; and

displaying, in the UI, at least a portion of the pre-build configuration.

20. The computer storage device of claim 19 , wherein the operations further comprise:

receiving, through the UI, an explainability setting selection;

displaying, in the UI, at least a portion of a hierarchy of the capacity prediction model; and

displaying, in the UI, information for a prior-existing computational resource used in generating the capacity prediction model.

Assignments (1)

Continuity (1)

References Cited (26)