Granted Patent B2

US 12,445,519 · App. 18/583,684 · Granted Oct 14, 2025

Inventors: Joseph W. Dain (Tucson, AZ); Frank N. Lee (Sunset Hills, MO); Yu-Cheng Hsu (Tucson, AZ)

Assignee: International Business Machines Corporation

H04L67/10H04L67/561H04L67/5651

Abstract

Claims (74)

1. A computer-implemented method (CIM), comprising:

receiving, at an edge computing environment, internet of things (IoT) data generated by IOT devices in communication with the edge computing environment;

storing the IOT data in a data layer at the edge computing environment in response to receiving the IOT data;

using, at the edge computing environment, one or more inference models to:

classify the IOT data, and

generate labeled metadata for the classified IOT data;

using, at the edge computing environment, the labeled metadata to identify sections of the IOT data to remove from the data layer;

removing the identified sections of the IOT data from the data layer;

using, at the edge computing environment, the labeled metadata to identify portions of the IOT data to transmit to a cloud computing environment; and

sending copies of the identified portions of the IOT data to the cloud computing environment.

2. The CIM of claim 1 , wherein the sending of the copies of the identified portions of the IOT data to the cloud computing environment includes:

sending one or more instructions to use the copies of the identified portions of the IOT data to train a source model that produced the one or more inference models.

3. The CIM of claim 2 , further comprising:

receiving, at the edge computing environment, an updated version of the inference models from the cloud computing environment; and

using the updated version of the inference models to:

classify new IOT data, and

generate labeled metadata for the classified new IOT data.

4. The CIM of claim 1 , further comprising:

indicating, in the data layer, that copies of the identified portions of the IOT data have been sent to the cloud computing environment.

5. The CIM of claim 1 , wherein the classifying of the IOT data includes:

applying one or more policies to metadata associated with the IOT data;

selecting portions of the IOT data that are highlighted as a result of applying the one or more policies; and

classifying the selected portions of the IOT data.

6. The CIM of claim 5 , wherein the generating of the labeled metadata includes:

generating labeled metadata for the selected portions of the IOT data.

7. The CIM of claim 1 , wherein the one or more inference models include trained AI models.

8. A computer program product (CPP), comprising:

a set of one or more computer-readable storage media; and

program instructions, collectively stored in the set of one or more storage media, for causing a processor set to perform the following computer operations:

receive, at an edge computing environment, internet of things (IoT) data generated by IOT devices in communication with the edge computing environment;

store the IOT data in a data layer at the edge computing environment in response to receiving the IOT data;

use, at the edge computing environment, one or more inference models to:

classify the IOT data, and

generate labeled metadata for the classified IOT data;

use, at the edge computing environment, the labeled metadata to identify sections of the IOT data to remove from the data layer;

remove the identified sections of the IOT data from the data layer;

use, at the edge computing environment, the labeled metadata to identify portions of the IOT data to transmit to a cloud computing environment; and

send copies of the identified portions of the IOT data to the cloud computing environment.

9. The CPP of claim 8 , wherein the sending of the copies of the identified portions of the IOT data to the cloud computing environment includes:

sending one or more instructions to use the copies of the identified portions of the IOT data to train a source model that produced the one or more inference models.

10. The CPP of claim 9 , wherein the program instructions are for causing the processor set to further perform the following computer operations:

receive, at the edge computing environment, an updated version of the inference models from the cloud computing environment; and

use the updated version of the inference models to:

classify new IOT data, and

generate labeled metadata for the classified new IOT data.

11. The CPP of claim 8 , wherein the program instructions are for causing the processor set to further perform the following computer operations:

indicate, in the data layer, that copies of the identified portions of the IOT data have been sent to the cloud computing environment.

12. The CPP of claim 8 , wherein the classifying of the IOT data includes:

applying one or more policies to metadata associated with the IOT data;

selecting portions of the IOT data that are highlighted as a result of applying the one or more policies; and

classifying the selected portions of the IOT data.

13. The CPP of claim 12 , wherein the generating of the labeled metadata includes:

generating labeled metadata for the selected portions of the IOT data.

14. The CPP of claim 8 , wherein the one or more inference models include trained AI models.

15. A computer system (CS), comprising:

a processor set;

a set of one or more computer-readable storage media;

program instructions, collectively stored in the set of one or more storage media, for causing the processor set to perform the following computer operations:

receive, at an edge computing environment, internet of things (IoT) data generated by IOT devices in communication with the edge computing environment;

store the IOT data in a data layer at the edge computing environment in response to receiving the IOT data;

use, at the edge computing environment, one or more inference models to:

classify the IOT data, and

generate labeled metadata for the classified IOT data;

use, at the edge computing environment, the labeled metadata to identify sections of the IOT data to remove from the data layer;

remove the identified sections of the IOT data from the data layer;

use, at the edge computing environment, the labeled metadata to identify portions of the IOT data to transmit to a cloud computing environment; and

send copies of the identified portions of the IOT data to the cloud computing environment.

16. The CS of claim 15 , wherein the program instructions are for causing the processor set to further perform the following computer operations:

indicate, in the data layer, that copies of the identified portions of the IOT data have been sent to the cloud computing environment.

17. The CS of claim 15 , wherein the classifying of the IOT data includes:

applying one or more policies to metadata associated with the IOT data;

selecting portions of the IOT data that are highlighted as a result of applying the one or more policies;

classifying the selected portions of the IOT data; and

generating labeled metadata for the selected portions of the IOT data.

Assignments (1)

Continuity (1)

References Cited (15)