Exploiting historical data: Pruning autotuning spaces and estimating the number of tuning steps

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14610%2F20%3A00116267" target="_blank" >RIV/00216224:14610/20:00116267 - isvavai.cz</a>

Výsledek na webu

<a href="https://onlinelibrary.wiley.com/doi/10.1002/cpe.5962" target="_blank" >https://onlinelibrary.wiley.com/doi/10.1002/cpe.5962</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1002/cpe.5962" target="_blank" >10.1002/cpe.5962</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Exploiting historical data: Pruning autotuning spaces and estimating the number of tuning steps

Popis výsledku v původním jazyce

Autotuning, the practice of automatic tuning of applications to provide performance portability, has received increased attention in the research community, especially in high performance computing. Ensuring high performance on a variety of hardware usually means modifications to the code, often via different values of a selected set of parameters, such as tiling size, loop unrolling factor, or data layout. However, the search space of all possible combinations of these parameters can be large, which can result in cases where the benefits of autotuning are outweighed by its cost, especially with dynamic tuning. Therefore, estimating the tuning time in advance or shortening the tuning time is very important in dynamic tuning applications. We have found that certain properties of tuning spaces do not vary much when hardware is changed. In this article, we demonstrate that it is possible to use historical data to reliably predict the number of tuning steps that is necessary to find a well-performing configuration and to reduce the size of the tuning space. We evaluate our hypotheses on a number of HPC benchmarks written in CUDA and OpenCL, using several different generations of GPUs and CPUs.

Název v anglickém jazyce

Exploiting historical data: Pruning autotuning spaces and estimating the number of tuning steps

Popis výsledku anglicky

Autotuning, the practice of automatic tuning of applications to provide performance portability, has received increased attention in the research community, especially in high performance computing. Ensuring high performance on a variety of hardware usually means modifications to the code, often via different values of a selected set of parameters, such as tiling size, loop unrolling factor, or data layout. However, the search space of all possible combinations of these parameters can be large, which can result in cases where the benefits of autotuning are outweighed by its cost, especially with dynamic tuning. Therefore, estimating the tuning time in advance or shortening the tuning time is very important in dynamic tuning applications. We have found that certain properties of tuning spaces do not vary much when hardware is changed. In this article, we demonstrate that it is possible to use historical data to reliably predict the number of tuning steps that is necessary to find a well-performing configuration and to reduce the size of the tuning space. We evaluate our hypotheses on a number of HPC benchmarks written in CUDA and OpenCL, using several different generations of GPUs and CPUs.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE

ISSN

1532-0626

e-ISSN

—

Svazek periodika

32

Číslo periodika v rámci svazku

21

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

15

Strana od-do

1-15

Kód UT WoS článku

000557422400001

EID výsledku v databázi Scopus

2-s2.0-85089146702