Energy-Efficient Near-Threshold Parallel Computing: The PULPv2 Cluster
- 11 October 2017
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 37 (5) , 20-31
Abstract
This article presents an ultra-low-power parallel computing platform and its system-on-chip (SoC) embodiment, targeting a wide range of emerging near-sensor processing tasks for Internet of Things (IoT) applications. The proposed SoC achieves 193 million operations per second (MOPS) per mW at 162 MOPS (32 bits), improving the first-generation Parallel Ultra-Low-Power (PULP) architecture by 6.4 and 3.2 times in performance and energy efficiency, respectively.Keywords
This publication has 10 references indexed in Scilit:
- Power, Area, and Performance Optimization of Standard Cell Memory Arrays Through Controlled PlacementACM Transactions on Design Automation of Electronic Systems, 2016
- 193 MOPS/mW @ 162 MOPS, 0.32V to 1.15V voltage range multi-core accelerator for energy efficient parallel and sequential digital processing2016
- A 60 GOPS/W, −1.8 V to 0.9 V body bias ULP cluster in 28 nm UTBB FD-SOI technologySolid-State Electronics, 2016
- 8.4 A 0.33V/-40°C process/temperature closed-loop compensation SoC embedding all-digital clock multiplier and DC-DC converter exploiting FDSOI 28nm back-gate biasing2015
- A 28 nm DSP Powered by an On-Chip LDO for High-Performance and Energy-Efficient Mobile ApplicationsIEEE Journal of Solid-State Circuits, 2014
- Ultra-wide body-bias range LDPC decoder in 28nm UTBB FDSOI technology2013
- Variation-Tolerant Architecture for Ultra Low Power Shared-L1 Processor ClustersIEEE Transactions on Circuits and Systems II: Express Briefs, 2012
- Fog computing and its role in the internet of things2012
- A 10 pJ/cycle ultra-low-voltage 32-bit microprocessor system-on-chip2011
- Near-Threshold Computing: Reclaiming Moore's Law Through Energy Efficient Integrated CircuitsProceedings of the IEEE, 2010