Monday
February 17, 2014
7:00pm - 9:00pm
Presentation:
Parallel Processing in R
presented by Dr. Norm Matloff, Computer Science Dept., UC Davis
The R programming language <http://www.r-project.org/> has become the
lingua franca of the data science world. It takes an object-oriented,
functional programming point of view, and has outstanding graphics
libraries. Best of all, it is open source.
R is not inherently parallel, lacking threads and message-passing
operations. However, due to a tremendous need for parallelism,
especially in this era of Big Data, a number of parallel libraries
have been developed. Dr. Matloff will discuss several in this talk,
one each for these platforms: clusters; multicore; cloud; and GPU.
(Most of these are OS-agnostic, but the multicore solutions run only
on Linux and other Unix-family systems.)
R is close enough to C that those without background in R should be
able to follow the presentation easily[*]. Similarly, though the talk
will assume a passing familiarity with threads, those lacking such
background should be able to follow the presentation as well.
[*] Dr. Matloff has an R Quick Start tutorial located here:
http://heather.cs.ucdavis.edu/~matloff/132/PLN/ProbStatBook.pdf