Reducing cache coherence traffic with a NUMA-aware runtime approach

Cache Coherent NUMA (ccNUMA) architectures are a widespread paradigm due to the benefits they provide for scaling core count and memory capacity. Also, the flat memory address space they offer considerably improves programmability. However, ccNUMA architectures require sophisticated and expensive ca...

Descripción completa

Detalles Bibliográficos
Autores: Caheny, Paul, Álvarez Martí, Lluc|||0000-0003-0506-8867, Derradji, Said, Valero Cortés, Mateo|||0000-0003-2917-2482, Moretó Planas, Miquel|||0000-0002-9848-8758, Casas, Marc|||0000-0003-4564-2093
Tipo de recurso: artículo
Fecha de publicación:2018
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:2117/116365
Acceso en línea:https://hdl.handle.net/2117/116365
https://dx.doi.org/10.1109/TPDS.2017.2787123
Access Level:acceso abierto
Palabra clave:Memory management (Computer science)
Parallel processing (Electronic computers)
Cache coherence
NUMA
Task-based programming models
Gestió de memòria (Informàtica)
Processament en paral·lel (Ordinadors)
Àrees temàtiques de la UPC::Informàtica::Arquitectura de computadors
Descripción
Sumario:Cache Coherent NUMA (ccNUMA) architectures are a widespread paradigm due to the benefits they provide for scaling core count and memory capacity. Also, the flat memory address space they offer considerably improves programmability. However, ccNUMA architectures require sophisticated and expensive cache coherence protocols to enforce correctness during parallel executions, which trigger a significant amount of on- and off-chip traffic in the system. This paper analyses how coherence traffic may be best constrained in a large, real ccNUMA platform comprising 288 cores through the use of a joint hardware/software approach. For several benchmarks, we study coherence traffic in detail under the influence of an added hierarchical cache layer in the directory protocol combined with runtime managed NUMA-aware scheduling and data allocation techniques to make most efficient use of the added hardware. The effectiveness of this joint approach is demonstrated by speedups of 3.14× to 9.97× and coherence traffic reductions of up to 99% in comparison to NUMA-oblivious scheduling and data allocation.