Random numbers take on a new significance in parallel programs, given that usually, you want to have different random number sequences in use across a set of cooperating parallel processes; simulation and optimum search type workloads being prime examples.
The default random number generator in R is Mersenne Twister and is generally recognized to be a good quality pseudorandom number generator, though it's not cryptographically very secure.
Note
Mersenne Twister
To find out more about the properties of the Mersenne Twister random number generator (RNG) you can refer to:
You can, of course, select alternate generators from the set of built-ins as well as supply your own using the base R random
package function RNGKind()
.
R in itself has always been a single-threaded implementation and is not designed to exploit parallelism within its own language primitives; it relies on specifically...