2012
DOI: 10.1155/2012/675130
|View full text |Cite
|
Sign up to set email alerts
|

A Hardware Efficient Random Number Generator for Nonuniform Distributions with Arbitrary Precision

Abstract: Nonuniform random numbers are key for many technical applications, and designing efficient hardware implementations of non-uniform random number generators is a very active research field. However, most state-of-the-art architectures are either tailored to specific distributions or use up a lot of hardware resources. At ReConFig 2010, we have presented a new design that saves up to 48% of area compared to state-of-the-art inversion-based implementation, usable for arbitrary distributions and precision. In this… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
5
0

Year Published

2013
2013
2016
2016

Publication Types

Select...
4
3
2

Relationship

1
8

Authors

Journals

citations
Cited by 15 publications
(5 citation statements)
references
References 17 publications
0
5
0
Order By: Relevance
“…Several methods for generating random number have been proposed indecent years. Most of the methods were implemented in hardware rather than software (De Schryver et al, 2012;VishnuRaj & Yuvaraj, 2014). These methods provide a maximum periodic of random sequence and high throughput rate while adhering to established statistical standards by applying a seeding mechanism.…”
Section: Related Workmentioning
confidence: 99%
“…Several methods for generating random number have been proposed indecent years. Most of the methods were implemented in hardware rather than software (De Schryver et al, 2012;VishnuRaj & Yuvaraj, 2014). These methods provide a maximum periodic of random sequence and high throughput rate while adhering to established statistical standards by applying a seeding mechanism.…”
Section: Related Workmentioning
confidence: 99%
“…The only drawback in our case is the cost of the mathematical operations involved; namely: sin, cos, log, sqrt, which are computationally intensive in all of the targeted platforms: CPU, XeonPhi, GPU, and FPGA. Other possibilities, such as the inverse cumulative distribution function (ICDF), have proven to be more resource-and energy-efficient [46,47] in specific architectures. However, we are interested in a portable project that we can deploy to fairly compare all targeted platforms.…”
Section: Local Random Number Generationmentioning
confidence: 99%
“…Fortunately, NVIDIA provides random generators in versions posterior to CUDA 3.2. In this work, the chosen random number generator is the Mersenne Twister method [26][27][28][29] available in the CUDA Software Development Kit [30]. This generator is invoked in the function RandomGPU, which generates random vector proportional to population size.…”
Section: Inputmentioning
confidence: 99%