Granted Patent B2

US 12,511,517 · App. 17/895,248 · Granted Dec 30, 2025

Inventors: Lalit Shinde (Thane, IN); Karthik Gowda (Navi Mumbai, IN); Chandan Singh (Bengaluru, IN); Abhinav Raj (Pune, IN); Rajesh Babu Madhavaraj (Chennai, IN); Surendravikram Singh (Bangalore, IN); Santhana Gopala Krishnan Natarajan (Frisco, TX)

Assignee: ACCENTURE GLOBAL SOLUTIONS LIMITED

G06N3/04G06F9/4875

Abstract

Claims (71)

1 . A computer-implemented method comprising:

receiving, by one or more computing devices, a plurality of parameters associated with an on-premises system to be migrated to a cloud architecture, the plurality of parameters including an identifier of the on-premises system, identifiers of components of the on-premises system, and migration requirements;

extracting, from the plurality of parameters, a set of input parameters substantially affecting a migration of the on-premises system to the cloud architecture,

wherein the set of input parameters comprises a database version and a number of central processing unit (CPU) cores, and

wherein the database version indicates whether support is available at a product vendor associated with the cloud architecture, or whether the database is outdated or on verge of support expiry;

identifying a target cloud architecture, selected from a plurality of cloud architectures, that i) is compliant with the set of input parameters, and ii) satisfies one or more threshold conditions associated with the migration;

determining, a set of output parameters representing features of the target cloud architecture; and

training, a neural network model using the set of input parameters and the set of output parameters, wherein training the neural network model comprises:

applying a first convolutional filter including a database version filter to generate a first feature map corresponding to the database version;

applying a second convolutional filter including a number of CPU calculation filter to generate a second feature map corresponding to the number of CPU cores;

combining the first feature map and the second feature map to generate a multi-dimensional filter; and

applying the multi-dimensional filter to each layer of a neural network of the neural network model.

2 . The computer-implemented method of claim 1 , wherein extracting the set of input parameters comprises:

executing extraction script at the on-premises system to extract metadata of the on-premises system; and

generating the set of input parameters based on analyzing the metadata of the on-premises system.

3 . The computer-implemented method of claim 1 ,

wherein the set of input parameters further comprises at least one of database edition, real application clusters (RAC), database size, database operating system, archive log mode, workload type, database services, autonomous database (ADB) features, database features, critical status, network bandwidth, and outage window, and

wherein the set of output parameters comprises at least one of: target cloud versions, migration methods, target cloud shapes and sizes, licensing impact, and estimated time.

4 . The computer-implemented method of claim 1 , wherein training the neural network model further comprises:

determining an order of multiple layers and a dependency between a node in a previous layer and a node in a layer following the previous layer in the neural network.

5 . The computer-implemented method of claim 1 , comprising:

after applying the multi-dimensional filter to each layer of the neural network to generate a result matrix; and

applying a rectified linear unit activation function to the result matrix to generate a second result matrix.

6 . The computer-implemented method of claim 5 , comprising:

applying max pooling to the second result matrix.

7 . The computer-implemented method of claim 1 , wherein the plurality of cloud architectures are from different vendors.

8 . The computer-implemented method of claim 1 , comprising;

receiving, by a server, a request to migrate a second on-premises system to a second cloud architecture, the request comprising a second plurality of parameters associated with the second on-premises system to be migrated to the second cloud architecture, the second plurality of parameters including an identifier of the second on-premises system, identifiers of components of the second on-premises system, and second migration requirements;

extracting, by the server from the second plurality of parameters, a second set of input parameters substantially affecting a migration of the second on-premises system to the second cloud architecture;

executing, by the server, the trained neural network model using the second set of input parameters to obtain a second set of output parameters associated with a second target cloud architecture; and

selecting, by the server from a plurality of cloud architectures, a second target cloud architecture using the second set of output parameters, wherein the second target cloud architecture i) is compliant with the second set of input parameters, and ii) satisfies one or more threshold conditions associated with the migration.

9 . A non-transitory computer-readable medium encoded with instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:

receiving a plurality of parameters associated with an on-premises system to be migrated to a cloud architecture, the plurality of parameters including an identifier of the on-premises system, identifiers of components of the on-premises system, and migration requirements;

extracting, from the plurality of parameters, a set of input parameters substantially affecting a migration of the on-premises system to the cloud architecture,

wherein the set of input parameters comprises a database version and a number of central processing unit (CPU) cores, and

wherein the database version indicates whether support is available at a product vendor associated with the cloud architecture, or whether the database is outdated or on verge of support expiry;

identifying a target cloud architecture, selected from a plurality of cloud architectures, that i) is compliant with the set of input parameters, and ii) satisfies one or more threshold conditions associated with the migration;

determining, a set of output parameters representing features of the target cloud architecture; and

training, a neural network model using the set of input parameters and the set of output parameters, wherein training the neural network model comprises:

applying a first convolutional filter including a database version filter to generate a first feature map corresponding to the database version;

applying a second convolutional filter including a number of CPU calculation filter to generate a second feature map corresponding to the number of CPU cores;

combining the first feature map and the second feature map to generate a multi-dimensional filter; and

applying the multi-dimensional filter to each layer of a neural network of the neural network model.

10 . The non-transitory computer-readable medium of claim 9 , wherein extracting the set of input parameters comprises:

executing extraction script at the on-premises system to extract metadata of the on-premises system; and

generating the set of input parameters based on analyzing the metadata of the on-premises system.

11 . The non-transitory computer-readable medium of claim 9 ,

wherein the set of input parameters further comprises at least one of database edition, real application clusters (RAC), database size, database operating system, archive log mode, workload type, database services, autonomous database (ADB) features, database features, critical status, network bandwidth, and outage window, and

wherein the set of output parameters comprises at least one of: target cloud versions, migration methods, target cloud shapes and sizes, licensing impact, and estimated time.

12 . The non-transitory computer-readable medium of claim 9 , wherein training the neural network model further comprises:

determining an order of multiple layers and a dependency between a node in a previous layer and a node in a layer following the previous layer in the neural network.

13 . A system comprising one or more computers and one or more storage devices on which are stored instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

receiving a plurality of parameters associated with an on-premises system to be migrated to a cloud architecture, the plurality of parameters including an identifier of the on-premises system, identifiers of components of the on-premises system, and migration requirements;

extracting, from the plurality of parameters, a set of input parameters substantially affecting a migration of the on-premises system to the cloud architecture,

wherein the set of input parameters comprises a database version and a number of central processing unit (CPU) cores, and

wherein the database version indicates whether support is available at a product vendor associated with the cloud architecture, or whether the database is outdated or on verge of support expiry;

identifying a target cloud architecture, selected from a plurality of cloud architectures, that i) is compliant with the set of input parameters, and ii) satisfies one or more threshold conditions associated with the migration;

determining, a set of output parameters representing features of the target cloud architecture; and

training, a neural network model using the set of input parameters and the set of output parameters, wherein training the neural network model comprises:

applying a first convolutional filter including a database version filter to generate a first feature map corresponding to the database version;

applying a second convolutional filter including a number of CPU calculation filter to generate a second feature map corresponding to the number of CPU cores;

combining the first feature map and the second feature map to generate a multi-dimensional filter; and

applying the multi-dimensional filter to each layer of a neural network of the neural network model.

14 . The system of claim 13 , wherein extracting the set of input parameters comprises:

executing extraction script at the on-premises system to extract metadata of the on-premises system; and

generating the set of input parameters based on analyzing the metadata of the on-premises system.

15 . The system of claim 13 ,

wherein the set of input parameters further comprises at least one of database edition, real application clusters (RAC), database size, database operating system, archive log mode, workload type, database services, autonomous database (ADB) features, database features, critical status, network bandwidth, and outage window, and

wherein the set of output parameters comprises at least one of: target cloud versions, migration methods, target cloud shapes and sizes, licensing impact, and estimated time.

16 . The system of claim 13 , wherein training the neural network model further comprises:

determining an order of multiple layers and a dependency between a node in a previous layer and a node in a layer following the previous layer in the neural network.

Assignments (2)

Continuity (1)

References Cited (10)