A Virtual Filter Based Fast Assessment methodology for Fault Tolerant NoCs

Abstract

Network on Chips (NoCs) as the compromising communication infrastructures in many-core system, are suffering the serious Multi-Cell Upsets (MCU) impacts. To accelerate the accurate assessment of the increasing and complex MCU in fault tolerant NoCs, we propose a virtual filter based analytical methodology. The approach converts an original MCU model to a new regenerated MCU model via fault tolerance filtering, and then maps the new MCU information into an existing boundary model to estimate the effectiveness of fault tolerant schemes fast. The diverse results demonstrate that the virtual filter based assessment achieves 473.75x speedup and only 11.454% accuracy loss over the latest fault injection.