Storage system with machine learning based skew prediction
US10430723B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Nov 23, 2015 |
| Grant date | Oct 1, 2019 |
| Priority date | — |
| Expiry date | Jun 27, 2038 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N5/01
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
An apparatus comprises a plurality of storage tiers, at least one data mover module, and skew predictor. A model generator processes information characterizing input-output activity involving one or more of the storage tiers in order to obtain skew measurements indicating portions of the input-output activity directed to portions of the one or more storage tiers for respective periods of time less than at least one corresponding time granularity, and generates a predictive model from the skew measurements. The skew predictor is configured in accordance with the predictive model to convert additional skew measurements obtained for a given period of time less than a desired time granularity to corresponding skew measurements in the desired time granularity. One or more of the converted skew measurements are utilized by the data mover module in controlling transfer of data between the storage tiers. The model generator is part of a machine learning system.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.