Accelerated Computing research group
A methodology for comparing optimization algorithms for auto-tuning

May 30, 2024 — As a result from the collaborations between different auto-tuning research groups that was kickstarted at the Lorentz Center Workshop in March 2022, our journal article on “A Methodology for Comparing Optimization Algorithms for Auto-Tuning” has now been published in Future Generation Computer Systems. The paper comes with an accompanying software package that makes it easy for researchers in auto-tuning to apply the new methodology.


Adapting applications to optimally utilize available hardware is no mean feat: the plethora of choices for optimization techniques are infeasible to maximize manually. To this end, auto-tuning frameworks are used to automate this task, which in turn use optimization algorithms to efficiently search the vast searchspaces. However, there is a lack of comparability in studies presenting advances in auto-tuning frameworks and the optimization algorithms incorporated. As each publication varies in the way experiments are conducted, metrics used, and results reported, comparing the performance of optimization algorithms among publications is infeasible. The auto-tuning community identified this as a key challenge at the 2022 Lorentz Center workshop on auto-tuning. The examination of the current state of the practice in this paper further underlines this. We propose a community-driven methodology composed of four steps regarding experimental setup, tuning budget, dealing with stochasticity, and quantifying performance. This methodology builds upon similar methodologies in other fields while taking into account the constraints and specific characteristics of the auto-tuning field, resulting in novel techniques. The methodology is demonstrated in a simple case study that compares the performance of several optimization algorithms used to auto-tune CUDA kernels on a set of modern GPUs. We provide a software tool to make the application of the methodology easy for authors, and simplifies reproducibility of results.


Floris-Jan Willemsen, Richard Schoonhoven, Jiří Filipovič, Jacob O. Tørring, Rob van Nieuwpoort, Ben van Werkhoven “A methodology for comparing optimization algorithms for auto-tuning” Future Generation Computer Systems (FGCS), 2024

Written by

Floris-Jan Willemsen

Floris-Jan is a PhD Candidate at Leiden University and the Netherlands eScience Center. His research focusses on intelligent, automated optimization of GPU software.