Granted Patent B2

US 11,983,573 · App. 17/376,249 · Granted May 14, 2024

Inventors: Eduardo Vera Sousa (Niterói, BR); Tiago Salviano Calmon (London, GB)

Assignee: EMC IP Holding Company LLC

G06F9/505G06F9/5083G06F11/3006G06F11/3433G06F18/22G06N3/04

Abstract

Claims (60)

1. A method for resource allocation using fingerprint representations of telemetry data, the method comprising:

receiving, at a resource allocation device, a request to execute a workload;

obtaining, by the resource allocation device, telemetry data associated with the workload;

identifying, by the resource allocation device, a breakpoint based on the telemetry data;

identifying, by the resource allocation device, a workload segment using the breakpoint;

generating, by the resource allocation device, a fingerprint representation using the workload segment;

performing, by the resource allocation device, a search in a fingerprint catalog using the fingerprint representation to identify a similar fingerprint;

obtaining, by the resource allocation device, a resource allocation policy associated with the similar fingerprint;

performing, by the resource allocation device, a resource policy application action based on the resource allocation policy;

identifying, by the resource allocation device, a second breakpoint based on the telemetry data;

identifying, by the resource allocation device, a second workload segment using the second breakpoint;

generating, by the resource allocation device, a second fingerprint representation using the second workload segment;

performing a second search in the fingerprint catalog using the second fingerprint representation to determine that no similar fingerprints are present in the fingerprint catalog;

decoding, based on the second search, the second fingerprint representation to obtain approximated telemetry data; and

obtaining a new resource allocation policy based on the approximated telemetry data.

2. The method of claim 1 , wherein the breakpoint is identified using an autoencoder neural network.

3. The method of claim 1 , wherein the fingerprint representation is generated using an autoencoder neural network.

4. The method of claim 1 , wherein the breakpoint is associated with a change in usage of a compute resource by the workload.

5. The method of claim 1 , wherein performing the search comprises using a distance function and a distance threshold to identify the similar fingerprint.

6. The method of claim 1 , wherein the decoding is performed using an autoencoder neural network.

7. A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for resource allocation using fingerprint representations of telemetry data, the method comprising:

receiving, at a resource allocation device, a request to execute a workload;

obtaining, by the resource allocation device, telemetry data associated with the workload;

identifying, by the resource allocation device, a breakpoint based on the telemetry data;

identifying, by the resource allocation device, a workload segment using the breakpoint;

generating, by the resource allocation device, a fingerprint representation using the workload segment;

performing, by the resource allocation device, a search in a fingerprint catalog using the fingerprint representation to identify a similar fingerprint;

obtaining, by the resource allocation device, a resource allocation policy associated with the similar fingerprint;

performing, by the resource allocation device, a resource policy application action based on the resource allocation policy;

identifying, by the resource allocation device, a second breakpoint based on the telemetry data;

identifying, by the resource allocation device, a second workload segment using the second breakpoint;

generating, by the resource allocation device, a second fingerprint representation using the second workload segment;

performing a second search in the fingerprint catalog using the second fingerprint representation to determine that no similar fingerprints are present in the fingerprint catalog;

decoding, based on the second search, the second fingerprint representation to obtain approximated telemetry data; and

obtaining a new resource allocation policy based on the approximated telemetry data.

8. The non-transitory computer readable medium of claim 7 , wherein the breakpoint is identified using an autoencoder neural network.

9. The non-transitory computer readable medium of claim 7 , wherein the fingerprint representation is generated using an autoencoder neural network.

10. The non-transitory computer readable medium of claim 7 , wherein the breakpoint is associated with a change in usage of a compute resource by the workload.

11. The non-transitory computer readable medium of claim 7 , wherein performing the search comprises using a distance function and a distance threshold to identify the similar fingerprint.

12. The non-transitory computer readable medium of claim 7 , wherein the decoding is performed using an autoencoder neural network.

13. A system for resource allocation using fingerprint representations of telemetry data, the system comprising:

a resource allocation device, executing on a processor comprising circuitry, and configured to:

receive a request to execute a workload;

obtain telemetry data associated with the workload;

identify a breakpoint based on the telemetry data;

identify a workload segment using the breakpoint;

generate a fingerprint representation using the workload segment;

perform a search in a fingerprint catalog using the fingerprint representation to identify a similar fingerprint;

obtain a resource allocation policy associated with the similar fingerprint;

perform a resource policy application action based on the resource allocation policy;

identify a second breakpoint based on the telemetry data;

identify a second workload segment using the second breakpoint;

generate a second fingerprint representation using the second workload segment;

perform a second search in the fingerprint catalog using the second fingerprint representation to determine that no similar fingerprints are present in the fingerprint catalog;

decode, based on the second search, the second fingerprint representation to obtain approximated telemetry data; and

obtain a new resource allocation policy based on the approximated telemetry data.

14. The system of claim 13 , wherein the breakpoint is identified using an autoencoder neural network.

15. The system of claim 13 , wherein the fingerprint representation is generated using an autoencoder neural network.

16. The system of claim 13 , wherein the breakpoint is associated with a change in usage of a compute resource by the workload.

17. The system of claim 13 , wherein performing the search comprises using a distance function and a distance threshold to identify the similar fingerprint.

Assignments (8)

Continuity (1)