On Revolution Analytics partner Cloudera's blog, Uri Laserson has posted an excellent guide to resampling from a large data set in Hadoop. Resampling is an important step in fitting ensemble models (including random forests and other bagging techniques), and Uri provides a step-by-step guide to implementing resampling methods using RHadoop. He provides the complete map-reduce code in the R language, as well as a useful script for installing RHadoop on a Cloudera instance.  

By the way, if you're new to RHadoop, here's RHadoop creator and project leader Antonio Piccolboni introducting RHadoop at last year's Strata CA conference.

  

Cloudera blog: How-to: Resample from a Large Data Set in Parallel (with R on Hadoop)