Understanding the thermal implications of multicore architectures

Multicore architectures are becoming the main design paradigm for current and future processors. The main reason is that multicore designs provide an effective way of overcoming instruction-level parallelism (ILP) limitations by exploiting thread-level parallelism (TLP). In addition, it is a power a...

Descripción completa

Detalles Bibliográficos
Autores: Chaparro, Pedro, González González, José, Magklis, Grigorios, Cai, Qiong, González Colás, Antonio María|||0000-0002-0009-0996
Tipo de recurso: artículo
Fecha de publicación:2007
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/101450
Acceso en línea:https://hdl.handle.net/2117/101450
https://dx.doi.org/10.1109/TPDS.2007.1092
Access Level:acceso abierto
Palabra clave:Microprocessors
Parallel processing (Electronic computers)
Multicore architectures
Dynamic thermal management
Activity migration
Dynamic voltage
Frequency scaling
Microprocessadors
Processament en paral·lel (Ordinadors)
Àrees temàtiques de la UPC::Informàtica::Arquitectura de computadors
Descripción
Sumario:Multicore architectures are becoming the main design paradigm for current and future processors. The main reason is that multicore designs provide an effective way of overcoming instruction-level parallelism (ILP) limitations by exploiting thread-level parallelism (TLP). In addition, it is a power and complexity-effective way of taking advantage of the huge number of transistors that can be integrated on a chip. On the other hand, today's higher than ever power densities have made temperature one of the main limitations of microprocessor evolution. Thermal management in multicore architectures is a fairly new area. Some works have addressed dynamic thermal management in bi/quad-core architectures. This work provides insight and explores different alternatives for thermal management in multicore architectures with 16 cores. Schemes employing both energy reduction and activity migration are explored and improvements for thread migration schemes are proposed.