MICRO Benchmark: Dynamic Viscosity

Benchmark from Tsoutsouras et al. MICRO paper¹ for comparison against the NIST Uncertainty Machine².

Calculates the dynamic viscosity of a sodium hydroxide solution in water by modeling the inputs as independent Gaussian variables.

Inputs

The samples are stored in a text file. The first line of the file is the number of samples that follow. See the source code to know which files are used and how.

Output

The application calculates a single output, the dynamic viscosity of the solution, selected with -S 0.

Example output using Signaloid's C0Pro-S core:

Building and Running Locally

Prerequisites

The native build needs GNU Make, a C compiler, and the GNU Scientific Library (GSL).

On macOS, install the Xcode Command Line Tools (which provide make and the C compiler) and then install GSL with Homebrew:

xcode-select --install
brew install gsl

On Linux (Debian/Ubuntu):

sudo apt-get install -y build-essential libgsl-dev

The top-level Makefile detects the GSL install location automatically, covering Homebrew on Apple Silicon (/opt/homebrew), Homebrew on Intel (/usr/local) and MacPorts (/opt/local), so no further configuration is needed on macOS.

From the repository root, run:

make local-build

This produces demo-native-mc at the repository root.

Run the resulting binary from the repository root, since it reads its samples-gaussian-dv_*.csv sample files by bare filename relative to the current working directory (symlinks to which exist at the repository root). For example, to run in Monte Carlo mode:

./demo-native-mc -S 0 -M 10000

Note that -M and -b both require -S. Using -M or -b without -S exits with a validation error (exit code 1).

---

References

Footnotes

1. Vasileios Tsoutsouras, Orestis Kaparounakis, Bilgesu Arif Bilgin, Chatura Samarakoon, James Timothy Meech, Jan Heck, Phillip Stanley-Marbell: The Laplace Microarchitecture for Tracking Data Uncertainty and Its Implementation in a RISC-V Processor. MICRO 2021: 1254-1269 ↩

2. Thomas Lafarge and Antonio Possolo. 2015. NIST Uncertainty Machine–User’s Manual. National Institute of Standards and Technology, Gaithersburg (2015). ↩