Schmidberger, Markus and Morgan, Martin and Eddelbuettel, Dirk and Yu, Hao and Tierney, Luke and Mansmann, Ulrich
State-of-the-Art in Parallel Computing with R.
Department of Statistics: Technical Reports, No.47
Journal of Statistical Software, 31
R is a mature open-source programming language for statistical computing and graphics. Many areas of statistical research are experiencing rapid growth in the size of data sets. Methodological advances drive increased use of simulations. A common approach is to use parallel computing. This paper presents an overview of techniques for parallel computing with R on computer clusters, on multi-core systems, and in grid computing. It reviews sixteen different packages, comparing them on their state of development, the parallel technology used, as well as on usability, acceptance, and performance. Two packages (snow, Rmpi) stand out as particularly useful for general use on computer clusters. Packages for grid computing are still in development, with only one package currently available to the end user. For multi-core systems four different packages exist, but a number of issues pose challenges to early adopters. The paper concludes with ideas for further developments in high performance computing with R. Example code is available in the appendix.