A tutorial on tidy cross-validation with R
Analyzing NetHack data, part 1: What kills the players
Analyzing NetHack data, part 2: What players kill the most
Building a shiny app to explore historical newspapers: a step-by-step guide
Classification of historical newspapers content: a tutorial combining R, bash and Vowpal Wabbit, part 1
Classification of historical newspapers content: a tutorial combining R, bash and Vowpal Wabbit, part 2
Dealing with heteroskedasticity; regression with robust standard errors using R
Easy time-series prediction with R: a tutorial with air traffic data from Lux Airport
Exporting editable plots from R to Powerpoint: making ggplot2 purrr with officer
Fast food, causality and R packages, part 1
Fast food, causality and R packages, part 2
For posterity: install {xml2} on GNU/Linux distros
Forecasting my weight with R
From webscraping data to releasing it as an R package to share with the world: a full tutorial with data from NetHack
Get text from pdfs or images using OCR: a tutorial with {tesseract} and {magick}
Getting data from pdfs using the pdftools package
Getting the data from the Luxembourguish elections out of Excel
Going from a human readable Excel file to a machine-readable csv with {tidyxl}
Historical newspaper scraping with {tesseract} and R
How Luxembourguish residents spend their time: a small {flexdashboard} demo using the Time use survey data
Imputing missing values in parallel using {furrr}
Looking into 19th century ads from a Luxembourguish newspaper with R
Making sense of the METS and ALTO XML standards
Manipulate dates easily with {lubridate}
Manipulating strings with the {stringr} package
Maps with pie charts on top of each administrative division: an example with Luxembourg's elections data
Missing data imputation and instrumental variables regression: the tidy approach
Objects types and some useful R functions for beginners
Pivoting data frames just got easier thanks to `pivot_wide()` and `pivot_long()`
R or Python? Why not both? Using Anaconda Python within R with {reticulate}
Searching for the optimal hyper-parameters of an ARIMA model in parallel: the tidy gridsearch approach
Some fun with {gganimate}
The best way to visit Luxembourguish castles is doing data science + combinatorial optimization
The never-ending editor war (?)
The year of the GNU+Linux desktop is upon us: using user ratings of Steam Play compatibility to play around with regex and the tidyverse
Using Data Science to read 10 years of Luxembourguish newspapers from the 19th century
Using a genetic algorithm for the hyperparameter optimization of a SARIMA model
Using the tidyverse for more than data manipulation: estimating pi with Monte Carlo methods
What hyper-parameters are, and what to do with them; an illustration with ridge regression
{pmice}, an experimental package for missing data imputation in parallel using {mice} and {furrr}
Building formulae
Functional peace of mind
Get basic summary statistics for all the variables in a data frame
Getting {sparklyr}, {h2o}, {rsparkling} to work together and some fun with bash
Importing 30GB of data into R with sparklyr
Introducing brotools
It's lists all the way down
It's lists all the way down, part 2: We need to go deeper
Keep trying that api call with purrr::possibly()
Lesser known dplyr 0.7* tricks
Lesser known dplyr tricks
Lesser known purrr tricks
Make ggplot2 purrr
Mapping a list of functions to a list of datasets with a list of columns as arguments
Predicting job search by training a random forest on an unbalanced dataset
Teaching the tidyverse to beginners
Why I find tidyeval useful
tidyr::spread() and dplyr::rename_at() in action
Easy peasy STATA-like marginal effects with R
Functional programming and unit testing for data munging with R available on Leanpub
How to use jailbreakr
My free book has a cover!
Work on lists of datasets instead of individual datasets by using functional programming
Method of Simulated Moments with R
New website!
Nonlinear Gmm with R - Example with a logistic regression
Simulated Maximum Likelihood with R
Bootstrapping standard errors for difference-in-differences estimation with R
Careful with tryCatch
Data frame columns as arguments to dplyr functions
Export R output to a file
I've started writing a 'book': Functional programming and unit testing for data munging with R
Introduction to programming econometrics with R
Merge a list of datasets together
Object Oriented Programming with R: An example with a Cournot duopoly
R, R with Atlas, R with OpenBLAS and Revolution R Open: which is fastest?
Read a lot of datasets at once with R
Unit testing with R
Update to Introduction to programming econometrics with R
Using R as a Computer Algebra System with Ryacas

In this short post, I benchmark different “versions” of R. I compare the execution speeds of R, R linked against OpenBLAS, R linked against ATLAS and Revolution R Open. Revolution R Open is a new open source version of R made by Revolution Analytics. It is linked against MKL and should offer huge speed improvements over vanilla R. Also, it uses every cores of your computer by default, without any change whatsoever to your code.

TL;DR: Revolution R Open is the fastest of all the benchmarked versions (with R linked against OpenBLAS and ATLAS just behind), and easier to setup.

I benchmarked these different versions of R using `R-benchmark-25.R`

that you can download here. This benchmark file was created by Simon Urbanek.

I ran the benchmarks on my OpenSUSE 13.2 computer with a Pentium Dual-Core CPU E6500@2.93GHz with 4GB of Ram. It's outdated, but it's still quite fast for most of my numerical computation needs. I installed “vanilla” R from the official OpenSUSE repositories which is currently at version 3.1.2.

Then, I downloaded OpenBLAS and ATLAS also from the official OpenSUSE repositories and made R use these libraries instead of its own implementation of BLAS. The way I did that is a bit hacky, but works: first, go to `/usr/lib64/R/lib`

and backup `libRblas.so`

(rename it to `libRblas.soBackup`

for instance). Then link `/usr/lib64/libopenblas.so.0`

to `/usr/lib64/R/lib/libRblas`

, and that's it, R will use OpenBLAS. For ATLAS, you can do it in the same fashion, but you'll find the library in `/usr/lib64/atlas/`

. These paths should be the same for any GNU/Linux distribution. For other operating systems, I'm sure you can find where these libraries are with Google.

The last version I benchmarked was Revolution R Open. This is a new version of R released by Revolution Analytics. Revolution Analytics had their own version of R, called Revolution R, for quite some time now. They decided to release a completely free as in freedom and free as in free beer version of this product which they now renamed Revolution R Open. You can download Revolution R Open here. You can have both “vanilla” R and Revolution R Open installed on your system.

I ran the `R-benchmark-25.R`

6 times for every version but will only discuss the 4 best runs.

R version | Fastest run | Slowest run | Mean Run |
---|---|---|---|

Vanilla R | 63.65 | 66.21 | 64.61 |

OpenBLAS R | 15.63 | 18.96 | 16.94 |

ATLAS R | 16.92 | 21.57 | 18.24 |

RRO | 14.96 | 16.08 | 15.49 |

As you can read from the table above, Revolution R Open was the fastest of the four versions, but not significantly faster than BLAS or ATLAS R. However, RRO uses all the available cores by default, so if your code relies on a lot matrix algebra, RRO might be actually a lot more faster than OpenBLAS and ATLAS R. Another advantage of RRO is that it is very easy to install, and also works with Rstudio and is compatible with every R package to existence. "Vanilla" R is much slower than the other three versions, more than 3 times as slow!

With other benchmarks, you could get other results, but I don't think that "vanilla" R could beat any of the other three versions. Whatever your choice, I recommend not using plain, “vanilla” R. The other options are much faster than standard R, and don't require much work to set up. I'd personally recommend Revolution R Open, as it is free software and compatible with CRAN packages and Rstudio.