ParallelR—Run Computationally Intensive
Tasks in R Quickly & Easily

REvolution's ParallelR, included as part of REvolution R Enterprise, provides parallelized routines for classification and bootstrapping statistical methods, as well as the ability to easily parallelize custom R programs.

 

Scale to Production with ParallelR

ParallelR—included with every edition of REvolution R Enterprise enables you to complete your R-based statistical analyses in a shorter period of time—taking full advantage of multiprocessor systems and network computers.

ParallelR Features

• Ever growing collection of R applications available
  simply by calling them.
• Migrate between platforms of varying scale without any
  changes to your code. Run ParallelR on the desktop, SMP
  server and multiple node clusters and grids - it's the same
  code, and there's no admin overhead!
• Easily migrate your code from sequential to parallel
  computing without extensive re-coding. ParallelR
  includes two packages, foreach and iterators that
  help you to perform loop parallelization quickly and easily.
  Simple foreach example. We offer a higher-level paradigm called “Sleigh” and a lower-level approach. Either way you will enjoy more programmer productivity and less system administration overhead than R-MPI provides.
• Avoid the difficult administration and initialization of multiple nodes in a cluster. ParallelR provides the tools to recruit nodes and parallelize complex programming tasks across the available nodes in the clusters, and reliably collect and collate the results from individual nodes.
• Intelligently schedule and guarantee the completion of batch workload across your distributed, High Performance Computing (HPC) environment. ParallelR is fully integrated with the job schedulers inside Platform Computing’s LSF, Sun’s Grid Engine, and Microsoft’s HPC Server 2008.
• Reliable fault-tolerant processing, which enables your calculation to complete even if one or more of your worker nodes goes offline during processing. ParallelR detects the fault, and automatically re-routes the remaining computations to the surviving worker nodes.
• Run on Mac, Windows or Linux

Contact us for more information and pricing on fully supported REvolution R Enterprise.

Have questions regarding any product or service? Contact our sales team for more information.