About: Offsite Autotuning Approach: Performance Model Driven Autotuning Applied to Parallel Explicit ODE Methods

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: Offsite Autotuning Approach: Performance Model Driven Autotuning Applied to Parallel Explicit ODE Methods Goto Sponge NotDistinct Permalink

An Entity of Type : schema:ScholarlyArticle, within Data Space : wasabi.inria.fr associated with source document(s)

Attributes	Values
type	Academic Article research paper schema:ScholarlyArticle
isDefinedBy	Covid-on-the-Web dataset
title	Offsite Autotuning Approach: Performance Model Driven Autotuning Applied to Parallel Explicit ODE Methods
Creator	Korch, Matthias Rauber, Thomas Seiferth, Johannes
source	PMC
abstract	Autotuning (AT) is a promising concept to minimize the often tedious manual effort of optimizing scientific applications for a specific target platform. Ideally, an AT approach can reliably identify the most efficient implementation variant(s) for a new platform or new characteristics of the input by applying suitable program transformations and analytic models. In this work, we introduce Offsite, an offline AT approach that automates this selection process at installation time by rating implementation variants based on an analytic performance model without requiring time-consuming runtime tests. From abstract multilevel description languages, Offsite automatically derives optimized, platform-specific and problem-specific code of possible variants and applies the performance model to these variants. We apply Offsite to parallel numerical methods for ordinary differential equations (ODEs). In particular, we investigate tuning a specific class of explicit ODE solvers, PIRK methods, for four different initial value problems (IVPs) on three different shared-memory systems. Our experiments demonstrate that Offsite can reliably identify the set of most efficient implementation variants for different given test configurations (ODE solver, IVP, platform) and effectively handle important AT scenarios.
has issue date	2020-05-22(xsd:dateTime)
bibo:doi	10.1007/978-3-030-50743-5_19
has license	cc-by
sha1sum (hex)	33b01ce50ca636e2c9a8789f6f7eefcb42c9a222
schema:url	https://doi.org/10.1007/978-3-030-50743-5_19
resource representing a document's title	Offsite Autotuning Approach: Performance Model Driven Autotuning Applied to Parallel Explicit ODE Methods
has PubMed Central identifier	PMC7295344
schema:publication	High Performance Computing
resource representing a document's body	covid:33b01ce50ca636e2c9a8789f6f7eefcb42c9a222#body_text
is schema:about of	named entity 'optimizing' named entity 'concept' named entity 'CONCEPT' named entity 'derives' named entity 'specific' named entity 'performance' named entity 'latency' named entity 'MPI' named entity 'shared-memory' named entity 'promising' named entity 'program transformations' named entity 'working set' named entity 'matrix multiplication' named entity 'load distribution' named entity 'computing' named entity 'ECM' named entity 'AMD' named entity 'ECM' named entity 'subset' named entity 'Runge-Kutta' named entity 'ECM' named entity 'subset' named entity 'cardinality' named entity 'black box' named entity 'use-case' named entity 'MPI' named entity 'semantic' named entity 'L3 cache' named entity 'time-consuming' named entity 'load/store' named entity 'ECM' named entity 'OpenMP' named entity 'Skylake' named entity 'ATF' named entity 'CPU' named entity 'program transformations' named entity 'OpenMP' named entity 'L1 data cache' named entity 'subset' named entity 'cluster systems' named entity 'constant propagation' named entity 'jitter' named entity 'previous systems' named entity 'subset' named entity 'subset' named entity 'numerical analysis' named entity 'memory hierarchy' named entity 'database' named entity 'pragmas' named entity 'vector operation' named entity 'computation time' named entity 'compiler' named entity 'optimized code' named entity 'database' named entity 'hyper-threading' named entity 'numerical algorithms' named entity 'OpenMP' named entity 'time intervals' named entity 'CPU' named entity 'arithmetic' named entity 'linear algebra' named entity 'program transformations' named entity 'scientific application' named entity 'ECM' named entity 'Variant Code' named entity 'electrical resistance' named entity 'differential evolution' named entity 'database' named entity 'compiler' named entity 'runtime behavior' named entity 'subset' named entity 'Intel C compiler' named entity 'automation' named entity 'memory latency' named entity 'maximums' named entity 'genetic algorithms' named entity 'cache levels' named entity 'Radau' named entity 'catastrophe model' named entity 'pragmas' named entity 'exhaustive search'

Faceted Search & Find service v1.13.91 as of Mar 24 2020

Alternative Linked Data Documents: Sponger | ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3229 as of Jul 10 2020, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (94 GB total memory)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2025 OpenLink Software