QA: Comparing RevoScaleR and MapReduce

How does the distributed computing functionality in RevoScaleR differ from MapReduce? 

Both RevoScaleR and MapReduce provide a framework for analyzing large data sets in R. RevoScaleR provides large data computing independent of whether you are working on a cluster; it contains external memory algorithms that process the data in chunks. Beginning with Revolution R Enterprise 5.0, RevoScaleR also provides a framework that allows one to use resources on an artibrary set of nodes on a cluster with any standard R function.   

Artikel-id: 3104267 - Laatst bijgewerkt: 29 okt. 2015 - Revisie: 1