Energy Efficient Network-on-Chip Architectures for Many-Core Near-Threshold Computing System

Abstract

Near threshold computing has unraveled a promising design space for energy efficient computing. However, it is still plagued by sub-optimal system performance. Application characteristics and hardware non-idealities of conventional architectures (those optimized for nominal voltage) prevent us from fully leveraging the potential of NTC systems. Increasing the computational core count still forms the bedrock of a multitude of contemporary works that address the problem of performance degradation in NTC systems. However, these works do not categorically address the shortcomings of the conventional on-chip interconnect fabric in a many core environment. In this work, we quantitatively demonstrate the performance bottleneck created by a conventional NTC architecture in many-core NTC systems. To reclaim the performance lost due to a sub-optimal NoC in many-core NTC systems, we propose BoostNoC-a power efficient, multi-layered network-on-chip architecture. BoostNoC improves the system performance by nearly 2× over a conventional NTC system, while largely sustaining its energy benefits. Further, capitalizing on the application characteristics, we propose two BoostNoC derivative designs: (i) PG BoostNoC; and (ii) Drowsy BoostNoC; to improve the energy efficiency by 1.4× and 1.37×, respectively over conventional NTC system.