Entropy transfers in the Linux Random Number Generator

François Goichon 1 Cédric Lauradoux 2 Guillaume Salagnac 1, 3 Thibaut Vuillemin 1
2 PLANETE - Protocols and applications for the Internet
Inria Grenoble - Rhône-Alpes, CRISAM - Inria Sophia Antipolis - Méditerranée
3 SOCRATE - Software and Cognitive radio for telecommunications
Inria Grenoble - Rhône-Alpes, CITI - CITI Centre of Innovation in Telecommunications and Integration of services
Abstract : One of the services provided by the operating system to the applications is random number generation. For security reasons, the Linux Random Number Generator is built upon the combination of a deterministic algorithm known as the cryptographic post-processing and an unpredictable physical phenomenon called an Entropy Source. While the various cryptographic post-processing algorithms and their properties are well described in the literature, the entropy collection process itself is little studied. This report first presents the different approaches to random number generation, and then details the architecture of the Linux Random Number Generator. Then, we present the experiments we performed to monitor entropy transfers. Our results show that the main source of randomness in the system is the behavior of the hard drive, and that most random numbers produced by the generator are actually consumed by the kernel itself.
Document type :
Reports
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/hal-00738638
Contributor : Cédric Lauradoux <>
Submitted on : Thursday, October 4, 2012 - 4:54:34 PM
Last modification on : Wednesday, December 19, 2018 - 10:00:05 AM
Long-term archiving on : Friday, December 16, 2016 - 9:10:19 PM

File

rr8060.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00738638, version 1

Collections

Citation

François Goichon, Cédric Lauradoux, Guillaume Salagnac, Thibaut Vuillemin. Entropy transfers in the Linux Random Number Generator. [Research Report] RR-8060, INRIA. 2012, pp.26. ⟨hal-00738638⟩

Share

Metrics

Record views

1178

Files downloads

2939