A fault-tolerant last level cache for CMPs operating at ultra-low voltage

Voltage scaling to values near the threshold voltage is a promising technique to hold off the many-core power wall. However, as voltage decreases, some SRAM cells are unable to operate reliably and show a behavior consistent with a hard fault. Block disabling is a micro-architectural technique that...

Descripción completa

Detalles Bibliográficos
Autores: Ferrerón, Alexandra, Alastruey, Jesús, Suárez Gracía, Dario, Monreal Arnal, Teresa|||0000-0002-0458-2234, Ibáñez Marín, Pablo Enrique, Viñals Yúfera, Víctor
Tipo de recurso: artículo
Fecha de publicación:2019
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:2117/127595
Acceso en línea:https://hdl.handle.net/2117/127595
https://dx.doi.org/10.1016/j.jpdc.2018.10.010
Access Level:acceso abierto
Palabra clave:Fault-tolerant computing
Multiprogramming (Electronic computers)
Near-threshold voltage
SRAM reliability
On-chip caches
Cache management
Tolerància als errors (Informàtica)
Multiprogramació (Ordinadors electrònics)
Àrees temàtiques de la UPC::Informàtica::Arquitectura de computadors
Descripción
Sumario:Voltage scaling to values near the threshold voltage is a promising technique to hold off the many-core power wall. However, as voltage decreases, some SRAM cells are unable to operate reliably and show a behavior consistent with a hard fault. Block disabling is a micro-architectural technique that allows low-voltage operation by deactivating faulty cache entries, at the expense of reducing the effective cache capacity. In the case of the last-level cache, this capacity reduction leads to an increase in off-chip memory accesses, diminishing the overall energy benefit of reducing the voltage supply. In this work, we exploit the reuse locality and the intrinsic redundancy of multi-level inclusive hierarchies to enhance the performance of block disabling with negligible cost. The proposed fault-aware last-level cache management policy maps critical blocks, those not present in private caches and with a higher probability of being reused, to active cache entries. Our evaluation shows that this fault-aware management results in up to 37.3% and 54.2% fewer misses per kilo instruction (MPKI) than block disabling for multiprogrammed and parallel workloads, respectively. This translates to performance enhancements of up to 13% and 34.6% for multiprogrammed and parallel workloads, respectively.