R: Use faster vecLib

These code snippets have been tested on R 3.1.0 and Mac OS 10.9.3. They presumably do *not* work on R 2.X!

## Enter these commands in the Mac OS Terminal

# use faster vecLib library
cd /Library/Frameworks/R.framework/Resources/lib
ln -sf  /System/Library/Frameworks/Accelerate.framework/Frameworks/vecLib.framework/Versions/Current/libBLAS.dylib libRblas.dylib

# return to default settings
cd /Library/Frameworks/R.framework/Resources/lib
ln -sf libRblas.0.dylib libRblas.dylib

# R script for benchmarking:
# http://r.research.att.com/benchmarks/R-benchmark-25.R
# On 2011 Macbook Pro (2.7 GHz Core i7): 43 sec. (Rblas) vs. 9 sec. (vecLib)

Do you know whether it's also possible to symlink the libLAPACK.dylib to R in place of the libRlapack.dylib?

I've tried:

cd /Library/Frameworks/R.framework/Resources/lib
mv libRlapack.dylib libRlapack.dylib.backup
ln -sf /System/Library/Frameworks/Accelerate.framework/Frameworks/vecLib.framework/Versions/Current/libLAPACK.dylib libRlapack.dylib

However R throws up an error upon launch.

I tried the same thing and it also failed.

However, I am wondering whether linking to BLAS only may be enough.

On my computer, the default BLAS from R gives:

> d <- 2e3
> system.time({ x <- matrix(rnorm(d^2),d,d); tcrossprod(x) })
   user  system elapsed 
  3.648   0.031   3.682 
> 
> d <- 1e3
> system.time({ x <- matrix(rnorm(d^2),d,d); svd(x) })
   user  system elapsed 
  3.400   0.020   3.422 

And veclib's BLAS gives:

> d <- 2e3
> system.time({ x <- matrix(rnorm(d^2),d,d); tcrossprod(x) })
   user  system elapsed 
  0.871   0.028   0.517 
> 
> d <- 1e3
> system.time({ x <- matrix(rnorm(d^2),d,d); svd(x) })
   user  system elapsed 
  1.049   0.091   0.569 

Assuming that the crossprod is from BLAS and the SVD from LAPACK, I wonder how to interpret the results above...

Actually, it seems that the brewed install points toward veclib's lapack:

> sessionInfo()
R version 3.4.1 (2017-06-30)
Platform: x86_64-apple-darwin16.6.0 (64-bit)
Running under: macOS Sierra 10.12.5

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libLAPACK.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
[1] compiler_3.4.1 tools_3.4.1   
> d <- 5e3
> system.time({ x <- matrix(rnorm(d^2),d,d); tcrossprod(x) })
   user  system elapsed 
 10.088   0.227   3.766 
> system.time({ x <- matrix(rnorm(d^2),d,d); solve(x) })
   user  system elapsed 
 24.642   0.425   6.205 

However, when using the standard R with linking to veclib's BLAS only, I obtain comparable performances:

> sessionInfo()
R version 3.4.0 (2017-04-21)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Sierra 10.12.5

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] tvmisc_0.1.0

loaded via a namespace (and not attached):
[1] compiler_3.4.0 tools_3.4.0    rstudioapi_0.6 fortunes_1.5-4
> d <- 5e3
> system.time({ x <- matrix(rnorm(d^2),d,d); tcrossprod(x) })
   user  system elapsed 
  9.464   0.243   3.568 
> system.time({ x <- matrix(rnorm(d^2),d,d); solve(x) })
   user  system elapsed 
 22.934   0.509   6.545