Data processing in distributed computing
US10268714B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Oct 30, 2015 |
| Grant date | Apr 23, 2019 |
| Priority date | — |
| Expiry date | Apr 6, 2037 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F16/2471
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
In an approach for processing a data block, a processor determines a first performance index of a first data node, wherein a first copy of the data block is stored on the first data node, wherein the first copy of the data block is a portion included in a data object. A processor determines a second performance index of a second data node, wherein a second copy of the data block is stored on the second data node. A processor processes the first copy of the data block using a MAP algorithm in response to the first performance index being greater than, according to a predefined threshold, the second performance index. A processor generates a processed result of the data object using a REDUCE algorithm based on, at least, the processed first copy of the data block.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.