Mth: Codesigned Hardware/Software Support for Fine Grain Threads

Multi-core processors are ubiquitous in all market segments from embedded to high performance computing, but only few applications can efficiently utilize them. Existing parallel frameworks aim to support thread-level parallelism in applications, but the imposed overhead prevents their usage for sma...

Descripción completa

Detalles Bibliográficos
Autor principal: Mocskos, Esteban Eduardo
Publicado: 2016
Materias:
Multicore processing
Multithreading
Parallel architectures
Parallel programming
Application programs
Graphic methods
Hardware
Hardware-software codesign
Multitasking
Parallel processing systems
Reconfigurable hardware
Ubiquitous computing
High performance computing
Multi-core processing
Multi-core processor
Multi-threading
Recursive algorithms
Software and hardwares
Synchronization systems
Thread level parallelism
Multicore programming
Acceso en línea:https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15566056_vPP_n99_p_GonzalezMarquez
http://hdl.handle.net/20.500.12110/paper_15566056_vPP_n99_p_GonzalezMarquez
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id paper:paper_15566056_vPP_n99_p_GonzalezMarquez
record_format dspace
spelling paper:paper_15566056_vPP_n99_p_GonzalezMarquez2023-06-08T16:23:14Z Mth: Codesigned Hardware/Software Support for Fine Grain Threads Mocskos, Esteban Eduardo Multicore processing Multithreading Parallel architectures Parallel programming Application programs Graphic methods Hardware Hardware-software codesign Multitasking Parallel architectures Parallel processing systems Parallel programming Reconfigurable hardware Ubiquitous computing High performance computing Multi-core processing Multi-core processor Multi-threading Recursive algorithms Software and hardwares Synchronization systems Thread level parallelism Multicore programming Multi-core processors are ubiquitous in all market segments from embedded to high performance computing, but only few applications can efficiently utilize them. Existing parallel frameworks aim to support thread-level parallelism in applications, but the imposed overhead prevents their usage for small problem instances. This work presents Micro-threads (Mth) a hardware-software proposal focused on a shared thread management model enabling the use of parallel resources in applications that have small chunks of parallel code or small problem inputs by a combination of software and hardware: delegation of the resource control to the application, an improved mechanism to store and fill processor's context, and an efficient synchronization system. Four sample applications are used to test our proposal: HSL filter (trivially parallel), FFT Radix2 (recursive algorithm), LU decomposition (barrier every cycle) and Dantzig algorithm (graph based, matrix manipulation). The results encourage the use of Mth and could smooth the use of multiple cores for applications that currently can not take advantage of the proliferation of the available parallel resources in each chip. © 2002-2011 IEEE. Fil:Mocskos, E. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. 2016 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15566056_vPP_n99_p_GonzalezMarquez http://hdl.handle.net/20.500.12110/paper_15566056_vPP_n99_p_GonzalezMarquez
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic Multicore processing
Multithreading
Parallel architectures
Parallel programming
Application programs
Graphic methods
Hardware
Hardware-software codesign
Multitasking
Parallel architectures
Parallel processing systems
Parallel programming
Reconfigurable hardware
Ubiquitous computing
High performance computing
Multi-core processing
Multi-core processor
Multi-threading
Recursive algorithms
Software and hardwares
Synchronization systems
Thread level parallelism
Multicore programming
spellingShingle Multicore processing
Multithreading
Parallel architectures
Parallel programming
Application programs
Graphic methods
Hardware
Hardware-software codesign
Multitasking
Parallel architectures
Parallel processing systems
Parallel programming
Reconfigurable hardware
Ubiquitous computing
High performance computing
Multi-core processing
Multi-core processor
Multi-threading
Recursive algorithms
Software and hardwares
Synchronization systems
Thread level parallelism
Multicore programming
Mocskos, Esteban Eduardo
Mth: Codesigned Hardware/Software Support for Fine Grain Threads
topic_facet Multicore processing
Multithreading
Parallel architectures
Parallel programming
Application programs
Graphic methods
Hardware
Hardware-software codesign
Multitasking
Parallel architectures
Parallel processing systems
Parallel programming
Reconfigurable hardware
Ubiquitous computing
High performance computing
Multi-core processing
Multi-core processor
Multi-threading
Recursive algorithms
Software and hardwares
Synchronization systems
Thread level parallelism
Multicore programming
description Multi-core processors are ubiquitous in all market segments from embedded to high performance computing, but only few applications can efficiently utilize them. Existing parallel frameworks aim to support thread-level parallelism in applications, but the imposed overhead prevents their usage for small problem instances. This work presents Micro-threads (Mth) a hardware-software proposal focused on a shared thread management model enabling the use of parallel resources in applications that have small chunks of parallel code or small problem inputs by a combination of software and hardware: delegation of the resource control to the application, an improved mechanism to store and fill processor's context, and an efficient synchronization system. Four sample applications are used to test our proposal: HSL filter (trivially parallel), FFT Radix2 (recursive algorithm), LU decomposition (barrier every cycle) and Dantzig algorithm (graph based, matrix manipulation). The results encourage the use of Mth and could smooth the use of multiple cores for applications that currently can not take advantage of the proliferation of the available parallel resources in each chip. © 2002-2011 IEEE.
author Mocskos, Esteban Eduardo
author_facet Mocskos, Esteban Eduardo
author_sort Mocskos, Esteban Eduardo
title Mth: Codesigned Hardware/Software Support for Fine Grain Threads
title_short Mth: Codesigned Hardware/Software Support for Fine Grain Threads
title_full Mth: Codesigned Hardware/Software Support for Fine Grain Threads
title_fullStr Mth: Codesigned Hardware/Software Support for Fine Grain Threads
title_full_unstemmed Mth: Codesigned Hardware/Software Support for Fine Grain Threads
title_sort mth: codesigned hardware/software support for fine grain threads
publishDate 2016
url https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15566056_vPP_n99_p_GonzalezMarquez
http://hdl.handle.net/20.500.12110/paper_15566056_vPP_n99_p_GonzalezMarquez
work_keys_str_mv AT mocskosestebaneduardo mthcodesignedhardwaresoftwaresupportforfinegrainthreads
_version_ 1768545897454501888

Ejemplares similares