Sie verwenden einen Browser, in dem JavaScript deaktiviert ist. Dadurch wird verhindert, dass Sie die volle Funktionalität dieser Webseite nutzen können. Zur Navigation müssen Sie daher die Sitemap nutzen.

You are currently using a browser with deactivated JavaScript. There you can't use all the features of this website. In order to navigate the site, please use the Sitemap .

An Investigation on Inherent Robustness of Posit Data Representation

Alouani, I. ,  BEN KHALIFA, A. ,  Merchant, F.Leupers, R.
Book Title:
Proceedings of the International Conference on VLSI Design (VLSID)
accepted for publication
Feb. 2021


As the dimensions and operating voltages of computer electronics shrink to cope with consumers’ demand for higher performance and lower power consumption, circuit sensitivity to soft errors increases dramatically. Recently, a new data-type is proposed in the literature called emph{posit} data type. Posit arithmetic has absolute advantages such as higher numerical accuracy, speed, and simpler hardware design than IEEE 754-2008 technical standard-compliant arithmetic. In this paper, we propose a comparative robustness study between 32-bit posit and 32-bit IEEE 754-2008 compliant representations. At first, we propose a theoretical analysis for IEEE 754 compliant numbers and posit numbers for single bit flip and double bit flips. Then, we conduct exhaustive fault injection experiments that show a considerable inherent resilience in posit format compared to classical IEEE 754 compliant representation. To show a relevant use-case of fault-tolerant applications, we perform experiments on a set of machine-learning applications.
In more than $95%$ of the exhaustive fault injection exploration, posit representation is less impacted by faults than the IEEE 754 compliant floating-point representation. Moreover, in $100%$ of the tested machine-learning applications, the accuracy of posit-implemented systems is higher than the classical floating-point-based ones.