Backup agent scaling with evaluation of prior backup jobs
US11481276B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Feb 13, 2020 |
| Grant date | Oct 25, 2022 |
| Priority date | — |
| Expiry date | Jan 21, 2041 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F2201/80
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A number of backup agents to be deployed to a system can be predicted by training one or more machine learning (ML) objects of a first prediction algorithm and training one or more ML objects of a second prediction algorithm. The training can be performed with archived backup job data. Both prediction algorithms can be applied to the backup job data to predict execution duration of the backup jobs. The prediction algorithm with a lower error can be used to predict a total execution duration of a current number of backup jobs. An optimal number of backup agents can be predicted based on the predicted total execution duration and the current number of backup jobs.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.