Fine Tuning Algorithmic Skeletons

Caromel, Denis; Leyton, Mario

doi:10.1007/978-3-540-74466-5_9

Cited by 31 publications

(25 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Java-based approaches Calcium [48], JaSkel [99], Lithium [10], Muskel [8], CO2P3S [152] and Skandium [142] are all implemented in Java. These frameworks mainly differ in their skeleton types and the skeleton composition techniques they imply.…”

Section: Skeleton Framework For Multicore Cpu Systems and Mpi-clustersmentioning

confidence: 99%

Performance-aware component composition for GPU-based systems

Dastgeer¹

2014

View full text Add to dashboard Cite

This thesis adresses issues associated with efficiently programming modern heterogeneous GPU-based systems, containing multicore CPUs and one or more programmable Graphics Processing Units (GPUs). We use ideas from component-based programming to address programming, performance and portability issues of these heterogeneous systems. Specifically, we present three approaches that all use the idea of having multiple implementations for each computation; performance is achieved/retained either a) by selecting a suitable implementation for each computation on a given platform or b) by dividing the computation work across different implementations running on CPU and GPU devices in parallel.In the first approach, we work on a skeleton programming library (SkePU) that provides high-level abstraction while making intelligent implementation selection decisions underneath either before or during the actual program execution. In the second approach, we develop a composition tool that parses extra information (metadata) from XML files, makes certain decisions offline, and, in the end, generates code for making the final decisions at runtime. The third approach is a framework that uses source-code annotations and program analysis to generate code for the runtime library to make the selection decision at runtime. With a generic performance modeling API alongside program analysis capabilities, it supports online tuning as well as complex program transformations.These approaches differ in terms of genericity, intrusiveness, capabilities and knowledge about the program source-code; however, they all demonstrate usefulness of component programming techniques for programming GPU-based systems. With experimental evaluation, we demonstrate how all three approaches, although different in their own way, provide good performance on different GPU-based systems for a variety of applications.This work has been supported by two EU FP7 projects (PEP-PHER, EXCESS) and by SeRC. Populärvetenskaplig sammanfattningAtt få varje generation av datorer att fungera snabbareär viktigt för samhä-llets utveckling och tillväxt. Traditionellt hade de flesta datorer bara en general-purpose processor (den så kallade CPU:n) som bara kunde exekvera en beräkningsuppgift i taget. Under det senasteårtiondet har dock flerkärniga och mångkärniga processorer blivit vanliga, och datorer har också blivit mer heterogena. Ett modernt datorsystem innehåller vanligtvis flerän en CPU, tillsammans med specialprocessorer såsom grafikprocessorer (GPU:er) som ar anpassade för att kunna exekvera vissa typer av beräkningar effektivarë an CPU:er. Vi kallar ett sådant system med en eller flera GPU:er för ett GPU-baserat system. GPU:er i sådana system har sitt eget separata minne, och för att kunna köra en beräkning på en GPU så behöver man vanligtvis flytta all indata till GPU:ns minne och sedan hämta tillbaka resultatet när beräkningenär klar.Programmeringen av GPU-baserade systemär icke-trivialt av flera anledningar: (1) CPU:er och GPU:er kräver olika programmeringsexper...

show abstract

Section: Skeleton Framework For Multicore Cpu Systems and Mpi-clustersmentioning

confidence: 99%

Performance-aware component composition for GPU-based systems

Dastgeer¹

2014

View full text Add to dashboard Cite

show abstract

“…As a skeleton library we use Skandium [2,3], which is a multi-core reimplementation of Calcium [4], a ProActive [5] based algorithmic skeleton library. Skandium is mainly inspired by Lithium [6] and Muskel [7] frameworks, developed at University of Pisa.…”

Section: Related Workmentioning

confidence: 99%

Exceptions for Algorithmic Skeletons

Leyton

Henrio

Piquer

2010

Euro-Par 2010 - Parallel Processing

Self Cite

View full text Add to dashboard Cite

Abstract. Algorithmic Skeletons offer high-level abstractions for parallel programming based on recurrent parallelism patterns. Patterns can be combined and nested into more complex parallelism behaviors. Programmers fill the skeleton patterns with the functional (business) code, which transforms the generic skeleton into a specific application. However, when the functional code generate exceptions, this exposes the programmer to details of the skeleton library, breaking the high-level abstraction principle. Furthermore, related parallel activities must be stopped as the exception is raised. This paper describes how to handle exceptions in Algorithmic Skeletons without breaking the high-level abstractions of the programming model. We describe both the behavior of the framework in a formal way, and its implementation in Java: the Skandium Library.

show abstract

“…Typical skeletons include both stream parallel and data parallel common patterns [7,8,9]. Programming frameworks based on algorithmic skeletons achieve a complete and useful separation of concerns between application programmers (in charge of recognizing parallelism exploitation patterns in the application at hand and of modeling them with suitable skeletons) and system programmers (in charge of solving, once and for all, during skeleton framework design and implementation, the problems related to the efficient implementation of skeletons and of skeleton composition).…”

Section: Background and Related Workmentioning

confidence: 99%

Stkm on Sca: A Unified Framework with Components, Workflows and Algorithmic Skeletons

Aldinucci

Bouziane

Danelutto

et al. 2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper investigates an implementation of STKM, a SpatioTemporal sKeleton Model. STKM expands the Grid Component Model (GCM) with an innovative programmable approach that allows programmers to compose an application by combining component, workflow and skeleton concepts. The paper deals with a projection of the STKM model on top of SCA and it evaluates its implementation using Tuscany Java SCA. Experimental results show the need and the benefits of the high level of abstraction offered by STKM.

show abstract

Fine Tuning Algorithmic Skeletons

Cited by 31 publications

References 7 publications

Performance-aware component composition for GPU-based systems

Performance-aware component composition for GPU-based systems

Exceptions for Algorithmic Skeletons

Stkm on Sca: A Unified Framework with Components, Workflows and Algorithmic Skeletons

Contact Info

Product

Resources

About