Fast normal random number generators on vector processors
For developers of high-performance computing applications requiring fast normal random number generation on vector processors, this paper provides practical implementation guidance and performance comparisons.
The paper implements vectorized Box-Muller and Polar methods for generating normal random numbers on vector processors, achieving good performance on the Fujitsu VP2200. It demonstrates that the Polar method outperforms other methods like the Ratio method and Von Neumann-Forsythe method on such architectures.
We consider pseudo-random number generators suitable for vector processors. In particular, we describe vectorised implementations of the Box-Muller and Polar methods, and show that they give good performance on the Fujitsu VP2200. We also consider some other popular methods, e.g. the Ratio method of Kinderman and Monahan (1977) (as improved by Leva (1992)), and the method of Von Neumann and Forsythe, and show why they are unlikely to be competitive with the Polar method on vector processors.