An Efficient Workload Distribution Mechanism for Tightly Coupled Heterogeneous Hardware

Ernesto Rivera-Alvarado, Francisco J. Torres-Rojas

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

Resumen

The Accelerated Processing Units (APUs) are tightly integrated heterogeneous computing platforms that contain a CPU and GPU in a single integrated circuit. As their internal computing resources use the same memory (RAM), they can efficiently share data structures. This opens up the opportunity for creating new types of specialized algorithms for the APU. In this research, we designed a specialized load distribution mechanism for the APU and used image rendering as our test load as it is a highly expensive computational task. We found out that our workload distribution mechanism provides better performance than conventional workload mechanisms for GPUs and multicore CPUs.

Idioma originalInglés
Título de la publicación alojadaIntelligent Sustainable Systems - Selected Papers of WorldS4 2022
EditoresAtulya K. Nagar, Dharm Singh Jat, Durgesh Kumar Mishra, Amit Joshi
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas243-251
Número de páginas9
ISBN (versión impresa)9789811976599
DOI
EstadoPublicada - 2023
Evento6th World Conference on Smart Trends in Systems, Security and Sustainability, WS4 2022 - London, Reino Unido
Duración: 24 ago 202227 ago 2022

Serie de la publicación

NombreLecture Notes in Networks and Systems
Volumen578
ISSN (versión impresa)2367-3370
ISSN (versión digital)2367-3389

Conferencia

Conferencia6th World Conference on Smart Trends in Systems, Security and Sustainability, WS4 2022
País/TerritorioReino Unido
CiudadLondon
Período24/08/2227/08/22

Huella

Profundice en los temas de investigación de 'An Efficient Workload Distribution Mechanism for Tightly Coupled Heterogeneous Hardware'. En conjunto forman una huella única.

Citar esto