Autotuning of OpenCL Kernels with Global Optimizations

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14610%2F17%3A00098442" target="_blank" >RIV/00216224:14610/17:00098442 - isvavai.cz</a>

Result on the web

<a href="https://dl.acm.org/citation.cfm?doid=3152821.3152877" target="_blank" >https://dl.acm.org/citation.cfm?doid=3152821.3152877</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3152821.3152877" target="_blank" >10.1145/3152821.3152877</a>

Result language

angličtina

Original language name

Autotuning of OpenCL Kernels with Global Optimizations

Original language description

Autotuning is an important method for automatically exploring code optimizations. It may target low-level code optimizations, such as memory blocking, loop unrolling or memory prefetching, as well as high-level optimizations, such as placement of computation kernels on proper hardware devices, optimizing memory transfers between nodes or between accelerators and main memory. In this paper, we introduce an autotuning method, which extends state-of-the-art low-level tuning of OpenCL or CUDA kernels towards more complex optimizations. More precisely, we introduce a Kernel Tuning Toolkit (KTT), which implements inter-kernel global optimizations, allowing to tune parameters affecting multiple kernels or also the host code. We demonstrate on practical examples, that with global kernel optimizations we are able to explore tuning options that are not possible if kernels are tuned separately. Moreover, our tuning strategies can take into account numerical accuracy across multiple kernel invocations and search for implementations within specific numerical error bounds.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

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

Project

<a href="/en/project/EF16_013%2F0001802" target="_blank" >EF16_013/0001802: CERIT Scientific Cloud</a><br>

Continuities

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

Publication year

2017

Confidentiality

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

Article name in the collection

1st Workshop on Autotuning and Adaptivity Approaches for Energy Efficient HPC Systems (ANDARE'2017)

ISBN

9781450353632

ISSN

—

e-ISSN

—

Number of pages

6

Pages from-to

—

Publisher name

ACM

Place of publication

Portland (USA)

Event location

Portland, Oregon, USA

Event date

Jan 1, 2017

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—