Using Batch Mode for Splus and R

When using R or S-plus for large simulations, it is adventageous to use the BATCH submission capabilities rather than opening an interactive session. There are many reasons for doing this, some of which are:

• you do not need to be logged into the computer post-submission,
• the cpu will dedicate more cycles to your job,
• etc...

Basically, if you are running a computationally intense simulation, run it in batch mode! Submitting a program in batch mode is done at the unix prompt. In R, simply type

pearson:~ > R-1.4.1 CMD BATCH batch.in batch.out

or in S

pearson:~ > splus6 BATCH batch.in batch2.out

where batch.in is a text file with R or S commands. An example batch file is given below.

x.bat <- rnorm(1000)
mean(x.bat)
var(x.bat)

Note that this file will generate 1000 standard normal variables and store this vector in x.bat. The mean and variance of this vector are then computed. Submitting this file to R will produce this outfile (similarly for S-plus):

R : Copyright 2002, The R Development Core Team
Version 1.4.1  (2002-01-30)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type `license()' or `licence()' for distribution details.

R is a collaborative project with many contributors.
Type `contributors()' for more information.

Type `demo()' for some demos, `help()' for on-line help, or
`help.start()' for a HTML browser interface to help.
Type `q()' to quit R.

> invisible(options(echo = TRUE))
> x.bat <- rnorm(1000)
> mean(x.bat)
[1] 0.002813524
> var(x.bat)
[1] 1.034795
> 
> 
> 
> proc.time()
[1] 0.70 0.12 1.92 0.00 0.00
>

It should be mentioned that the S or R object x.bat will be in the current S or R working directory. That is, when you start S or R within the same directory, you will see x.bat if you type ls(). The advantage of running a S-plus or R in batch mode is not made apparent by the example given above. However, as your programs become increasingly complex, batch mode might be the way to go.