Longevity based computer resource provisioning

Patent metadata

Field	Value
Publication number	US-11263035-B2
Application number	US-201815953131-A
Country	US
Kind code	B2
Filing date	Apr 13, 2018
Priority date	Apr 13, 2018
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

Computer resources are provisioned for a virtual machine based on expected lifespan. After a request to create a virtual machine is received, the virtual machine can be classified into one of a plurality of longevity classes utilizing a machine learning classifier based on data pertaining to the requestor or the virtual machine. The longevity classes capture different lifespans of the virtual machine between when the virtual machine is created and when the virtual machine is deleted. Subsequently, resources for the virtual machine are provisioned from a hardware resource pool of a set of disjoint resource pools specific to the longevity class of the virtual machine.

First claim

What is claimed is:

1. A computer resource provisioning system, comprising:
a processor coupled to a memory, the processor configured to execute computer-executable instructions stored in the memory that when executed cause the processor to perform the following actions:
training a machine learning classifier to classify databases into a plurality of longevity classes based on a set of features that includes at least a number or rate of distinct characters included in names of the databases, wherein the longevity classes capture different database lifespans and wherein a database lifespan represents a length of time between when a database is created and subsequently deleted;
detecting a request to create a database from a requestor, the request including at least a name for the requested database;
classifying the requested database into one of the plurality of longevity classes with the machine learning classifier based at least on the name for the requested database; and
provisioning resources for the requested database from a hardware resource pool of a set of disjoint resource pools, wherein the hardware resource pool is specific to the longevity class of the requested database.

2. The system of claim 1 , wherein the machine learning classifier generates a predictive score and the requested database is classified into one of the plurality of longevity classes based on comparison of the predictive score to a threshold.

3. The system of claim 1 , wherein the plurality of longevity classes comprises a long-lived class and a short-lived class.

4. The system of claim 1 , wherein the actions further comprise:
detecting a misclassification of a database based on lifespan; and
updating the machine learning classifier automatically with the misclassification.

5. The system of claim 4 , wherein the actions further comprise migrating resources of a misclassified database from a first resource pool to a second resource pool, wherein the second resource pool corresponds to a new class based on a current lifespan.

6. The system of claim 1 wherein the requested database is a database provided by a cloud database service.

7. The system of claim 1 wherein the classifying further comprises classifying the requested database based on historical actions of the requestor.

8. The system of claim 1 wherein the classifying further comprises classifying the requested database based on a size of the requested database.

9. The system of claim 1 , wherein the machine learning classifier implements a random forest model to classify databases.

10. A method of resource provisioning comprising:
training a machine learning classifier to classify databases into a plurality of longevity classes based on a set of features that includes at least a presence of non-alphanumeric symbols included in names of the databases, wherein the longevity classes capture different database lifespans and wherein a database lifespan represents a length of time between when a database is created and subsequently deleted;
detecting a request to create a database from a requestor, the request including at least a name for the requested database;
classifying the database into one of the plurality of longevity classes with the machine learning classifier based at least on the name for the requested database; and
provisioning resources for the requested database from a hardware resource pool from a set of disjoint resource pools, wherein the hardware resource pool is specific to the longevity class of the requested database.

11. The method of claim 10 wherein the classifying is based on a comparison of a predictive score produced by the machine learning classifier to a predetermined threshold.

12. The method of claim 10 wherein the plurality of longevity classes comprises a short-lived class and a long-lived class.

13. The method of claim 10 further comprising:
detecting a misclassification of a database based on its current lifespan; and
updating the machine learning classifier automatically based on the misclassification.

14. The method of claim 13 further comprising moving a misclassified database to a different longevity class corresponding with its current lifespan.

15. The method of claim 10 wherein the requested database is a database provided by a cloud database service.

16. The method of claim 10 wherein the classifying is based on at least one of historical actions of the requestor or size of the requested database provided by the database.

17. A system for provisioning computer resources, comprising:
a classifier training system for training a machine learning classifier to classify databases into a plurality of longevity classes based on a set of features that includes a number and rate of distinct characters included in names of the databases, wherein the longevity classes capture different database lifespans and wherein a database lifespan represents a length of time between when a database is created and subsequently deleted;
means for classifying a requested database into one of the plurality of longevity classes with the machine learning classifier based on identity of a requestor that requests creation of the requested database and features of the database including at least a name of the requested database specified by the requestor; and
means for provisioning resources for the requested database from a hardware resource pool from a set of disjoint resource pools, wherein the hardware resource pool is designated for the longevity class of the requested database.

18. The system of claim 17 , wherein the machine learning classifier further produces a predictive score and the requested database is classified into one of the plurality of longevity classes based on comparison the predictive score to a predetermined threshold.

19. The system of claim 18 , wherein the machine learning classifier produces the predictive score further based on at least one of creation time or size of the requested database.

20. The system of claim 17 further comprising:
means for detecting misclassification of a database; and
means for migrating database resources from a first resource pool to a second resource pool corresponding to a new class based on a current lifespan of the misclassified database.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Lang Willis
• Picado Leiva Jose Manuel

Classifications

• G06F9/5077

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F9/5061 Primary

  Partitioning or combining of resources · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F2009/4557

  Distribution of virtual machine instances; Migration and load balancing · CPC title