Patent · US Active

Systems and methods for conserving computational resources by reducing network traffic predictions via machine learning models

US12407627B2 · kind B2 · utility

0Cited by
4References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJan 26, 2024
Grant dateSep 2, 2025
Priority date
Expiry dateJan 26, 2044

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH04L47/801
  • WIPO fieldDigital communication
  • WIPO sectorElectrical engineering

Abstract

In some embodiments, conserving computational resources by reducing network traffic predictions via machine learning models may be facilitated. In some embodiments, in connection with a first scheduled time for automated triggering of network resource predictions via a first machine learning model at scheduled times, a first resource usage prediction is received, via the first machine learning model, for a service executing within a computing environment. A first resource allocation process is performed for the service based on the first resource usage prediction. In connection with detecting that an amount of actual traffic corresponding to the service satisfies a traffic threshold for a threshold amount of time, a second resource allocation process is performed, before a second scheduled time next in the scheduled times after the first scheduled time.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.