Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications

Eduard Ayguade; Rosa M. Badia; Pieter Bellens; Javier Bueno; Alejandro Duran; Yoav Etsion; Montse Farreras; Roger Ferrer; Jesus Labarta; Vladimir Marjanovic; Lluis Martinell; Xavier Martorell; Josep M. Perez; Judit Planas; Alex Ramirez; Xavier Teruel; Ioanna Tsalouchidou; Mateo Valero

doi:10.1002/9781119332015.ch5

Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications

Eduard Ayguade, Rosa M. Badia, Pieter Bellens, Javier Bueno, Alejandro Duran, Yoav Etsion, Montse Farreras, Roger Ferrer, Jesus Labarta, Vladimir Marjanovic, Lluis Martinell, Xavier Martorell, Josep M. Perez, Judit Planas, Alex Ramirez, Xavier Teruel, Ioanna Tsalouchidou, Mateo Valero

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

Abstract

This chapter describes how OmpSs extends the OpenMP 3.0 node programming model and how it leverages message passing interface (MPI) and OpenCL/CUDA, mastering the efficient programming of the clustered heterogeneous multi-/many-core systems that will be available in current and future computing systems. It describes the language extensions and the implementation of OmpSs, focusing on the intelligence that needs to be embedded in the runtime system to effectively lower the programmability wall and the opportunities to implement new mechanisms and policies. The chapter reasons about the overheads related with task management (detecting intertask data dependencies, identifying task-level parallelism and executing tasks out of order) in OmpSs examining how far a software implementation can go to cope with fine-grain parallelism and opening the door to novel hardware mechanisms for emerging multicore architectures. The chapter provides a brief description of the OmpSs execution model to understand the programming model extensions.

Original language	English
Title of host publication	Programming Multicore and Many-Core Computing Systems
Pages	99-120
Number of pages	22
ISBN (Electronic)	9781119332015
DOIs	https://doi.org/10.1002/9781119332015.ch5
State	Published - 27 Jan 2017
Externally published	Yes

Keywords

Clustered heterogeneous multi-/many-core systems
Computing systems
Message passing interface
Multicore architectures
OmpSs execution model
OpenCL/CUDA
OpenMP 3.0 node programming model
Programming model extensions
Task management

All Science Journal Classification (ASJC) codes

General Computer Science

Access to Document

10.1002/9781119332015.ch5

Cite this

Ayguade, E., Badia, R. M., Bellens, P., Bueno, J., Duran, A., Etsion, Y., Farreras, M., Ferrer, R., Labarta, J., Marjanovic, V., Martinell, L., Martorell, X., Perez, J. M., Planas, J., Ramirez, A., Teruel, X., Tsalouchidou, I., & Valero, M. (2017). Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications. In Programming Multicore and Many-Core Computing Systems (pp. 99-120) https://doi.org/10.1002/9781119332015.ch5

@inbook{3407592323794b8788593cf98942b65a,

title = "Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications",

abstract = "This chapter describes how OmpSs extends the OpenMP 3.0 node programming model and how it leverages message passing interface (MPI) and OpenCL/CUDA, mastering the efficient programming of the clustered heterogeneous multi-/many-core systems that will be available in current and future computing systems. It describes the language extensions and the implementation of OmpSs, focusing on the intelligence that needs to be embedded in the runtime system to effectively lower the programmability wall and the opportunities to implement new mechanisms and policies. The chapter reasons about the overheads related with task management (detecting intertask data dependencies, identifying task-level parallelism and executing tasks out of order) in OmpSs examining how far a software implementation can go to cope with fine-grain parallelism and opening the door to novel hardware mechanisms for emerging multicore architectures. The chapter provides a brief description of the OmpSs execution model to understand the programming model extensions.",

keywords = "Clustered heterogeneous multi-/many-core systems, Computing systems, Message passing interface, Multicore architectures, OmpSs execution model, OpenCL/CUDA, OpenMP 3.0 node programming model, Programming model extensions, Task management",

author = "Eduard Ayguade and Badia, \{Rosa M.\} and Pieter Bellens and Javier Bueno and Alejandro Duran and Yoav Etsion and Montse Farreras and Roger Ferrer and Jesus Labarta and Vladimir Marjanovic and Lluis Martinell and Xavier Martorell and Perez, \{Josep M.\} and Judit Planas and Alex Ramirez and Xavier Teruel and Ioanna Tsalouchidou and Mateo Valero",

year = "2017",

month = jan,

day = "27",

doi = "10.1002/9781119332015.ch5",

language = "الإنجليزيّة",

isbn = "9780470936900",

pages = "99--120",

booktitle = "Programming Multicore and Many-Core Computing Systems",

}

Ayguade, E, Badia, RM, Bellens, P, Bueno, J, Duran, A, Etsion, Y, Farreras, M, Ferrer, R, Labarta, J, Marjanovic, V, Martinell, L, Martorell, X, Perez, JM, Planas, J, Ramirez, A, Teruel, X, Tsalouchidou, I & Valero, M 2017, Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications. in Programming Multicore and Many-Core Computing Systems. pp. 99-120. https://doi.org/10.1002/9781119332015.ch5

TY - CHAP

T1 - Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications

AU - Ayguade, Eduard

AU - Badia, Rosa M.

AU - Bellens, Pieter

AU - Bueno, Javier

AU - Duran, Alejandro

AU - Etsion, Yoav

AU - Farreras, Montse

AU - Ferrer, Roger

AU - Labarta, Jesus

AU - Marjanovic, Vladimir

AU - Martinell, Lluis

AU - Martorell, Xavier

AU - Perez, Josep M.

AU - Planas, Judit

AU - Ramirez, Alex

AU - Teruel, Xavier

AU - Tsalouchidou, Ioanna

AU - Valero, Mateo

PY - 2017/1/27

Y1 - 2017/1/27

N2 - This chapter describes how OmpSs extends the OpenMP 3.0 node programming model and how it leverages message passing interface (MPI) and OpenCL/CUDA, mastering the efficient programming of the clustered heterogeneous multi-/many-core systems that will be available in current and future computing systems. It describes the language extensions and the implementation of OmpSs, focusing on the intelligence that needs to be embedded in the runtime system to effectively lower the programmability wall and the opportunities to implement new mechanisms and policies. The chapter reasons about the overheads related with task management (detecting intertask data dependencies, identifying task-level parallelism and executing tasks out of order) in OmpSs examining how far a software implementation can go to cope with fine-grain parallelism and opening the door to novel hardware mechanisms for emerging multicore architectures. The chapter provides a brief description of the OmpSs execution model to understand the programming model extensions.

AB - This chapter describes how OmpSs extends the OpenMP 3.0 node programming model and how it leverages message passing interface (MPI) and OpenCL/CUDA, mastering the efficient programming of the clustered heterogeneous multi-/many-core systems that will be available in current and future computing systems. It describes the language extensions and the implementation of OmpSs, focusing on the intelligence that needs to be embedded in the runtime system to effectively lower the programmability wall and the opportunities to implement new mechanisms and policies. The chapter reasons about the overheads related with task management (detecting intertask data dependencies, identifying task-level parallelism and executing tasks out of order) in OmpSs examining how far a software implementation can go to cope with fine-grain parallelism and opening the door to novel hardware mechanisms for emerging multicore architectures. The chapter provides a brief description of the OmpSs execution model to understand the programming model extensions.

KW - Clustered heterogeneous multi-/many-core systems

KW - Computing systems

KW - Message passing interface

KW - Multicore architectures

KW - OmpSs execution model

KW - OpenCL/CUDA

KW - OpenMP 3.0 node programming model

KW - Programming model extensions

KW - Task management

UR - http://www.scopus.com/inward/record.url?scp=85019588601&partnerID=8YFLogxK

U2 - 10.1002/9781119332015.ch5

DO - 10.1002/9781119332015.ch5

M3 - فصل

SN - 9780470936900

SP - 99

EP - 120

BT - Programming Multicore and Many-Core Computing Systems

ER -

Hybrid/Heterogeneous Programming with OmpSs and its Software/Hardware Implications

Abstract

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this