Figure 1.

MapReduce Framework. MapReduce provides a generic framework that enables rapid, parallel analysis of partitionable data. Each Mapper runs a single test (e.g. OutputCoverage) across a section of the data from a single sequence file. The Reduce phase (if specified) can do additional analysis across aggregated data or mapper results and output information that can be used by external tools (as in Figure 2).

Robinson et al. BMC Genomics 2011 12:419   doi:10.1186/1471-2164-12-419
Download authors' original image