Distributing data analytics in a hierarchical network based on computational complexity
US10291494B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Apr 20, 2016 |
| Grant date | May 14, 2019 |
| Priority date | — |
| Expiry date | May 23, 2037 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04L41/14
- WIPO fieldDigital communication
- WIPO sectorElectrical engineering
Abstract
A method provided in a network including edge devices to collect data from data producers connected to the edge devices and to communicate with cloud-based prosumers connected with the edge devices. Data analytics tasks are identified. The data analytics tasks are used to process data collected from a data producer among the data producers to produce a result for consumption by one or more of the cloud-based prosumers. For each data analytics task it is determined whether a computational complexity of the data analytics task is less than or equal to a predetermined computational complexity. Each data analytics task determined to have a computational complexity less than or equal to the predetermined computational complexity is assigned to an edge device among the edge devices. Each data analytics task determined to have a computational complexity that exceeds the predetermined computational complexity is assigned to a prosumer among the prosumers.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.