Search engine for discovering works of Art, research articles, and books related to Art and Culture
ShareThis
Javascript must be enabled to continue!

Application-independent Autotuning for GPUs

View through CrossRef
Autotuning is an established technique for adjusting performance-critical parameters of applications to their specific run-time environment. In this paper, we investigate the potential of online autotuning for general purpose computation on GPUs. Our application-independent autotuner AtuneRT optimizes GPU-specific parameters such as block size and loop-unrolling degree. We also discuss the peculiarities of autotuning on GPUs. We demonstrate tuning potential using CUDA and by instrumenting the parallel algorithms library Thrust. We evaluate our online autotuning approach with various GPUs and sample applications.
Title: Application-independent Autotuning for GPUs
Description:
Autotuning is an established technique for adjusting performance-critical parameters of applications to their specific run-time environment.
In this paper, we investigate the potential of online autotuning for general purpose computation on GPUs.
Our application-independent autotuner AtuneRT optimizes GPU-specific parameters such as block size and loop-unrolling degree.
We also discuss the peculiarities of autotuning on GPUs.
We demonstrate tuning potential using CUDA and by instrumenting the parallel algorithms library Thrust.
We evaluate our online autotuning approach with various GPUs and sample applications.

Related Results

On the programmability of multi-GPU computing systems
On the programmability of multi-GPU computing systems
Multi-GPU systems are widely used in High Performance Computing environments to accelerate scientific computations. This trend is expected to continue as integrated GPUs will be i...
Autotuning PolyBench benchmarks with LLVM Clang/Polly loop optimization pragmas using Bayesian optimization
Autotuning PolyBench benchmarks with LLVM Clang/Polly loop optimization pragmas using Bayesian optimization
AbstractWe develop a ytopt autotuning framework that leverages Bayesian optimization to explore the parameter space search and compare four different supervised learning methods wi...
autotuning with machine learning of OpenMP task applications
autotuning with machine learning of OpenMP task applications
Autotuning assisté par apprentissage automatique de tâches OpenMP Les architectures informatiques modernes sont très complexes, nécessitant un grand effort de progr...
Autotuning divide‐and‐conquer stencil computations
Autotuning divide‐and‐conquer stencil computations
SummaryThis paper explores autotuning strategies for serial divide‐and‐conquer stencil computations, comparing the efficacy of traditional “heuristic” autotuning with that of “prun...
Toward transparent and parsimonious methods for automatic performance tuning
Toward transparent and parsimonious methods for automatic performance tuning
Vers des méthodes transparentes et parcimonieuses pour l'optimisation automatique des performances La fin de la loi de Moore et de la loi de Dennard entraînent une ...
Résolution de systèmes linéaires et non linéaires creux sur grappes de GPUs
Résolution de systèmes linéaires et non linéaires creux sur grappes de GPUs
Depuis quelques années, les grappes équipées de processeurs graphiques GPUs sont devenues des outils très attrayants pour le calcul parallèle haute performance. Dans cette thèse, n...
Investigating and Reducing the Architectural Impact of Transient Faults in Special Function Units for GPUs
Investigating and Reducing the Architectural Impact of Transient Faults in Special Function Units for GPUs
AbstractEnsuring the reliability of GPUs and their internal components is paramount, especially in safety-critical domains like autonomous machines and self-driving cars. These cut...

Back to Top