research article
Energy-Efficient Near-Threshold Parallel Computing: The PULPv2 Cluster
2017
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.
Type
research article
Web of Science ID
WOS:000413024800004
Author(s)
Rossi, Davide
Pullini, Antonio
Loi, Igor
Gautschi, Michael
Gurkaynak, Frank Kagan
Teman, Adam
Constantin, Jeremy
Miro-Panades, Ivan
Beigne, Edith
Date Issued
2017
Published in
Volume
37
Issue
5
Start page
20
End page
31
Editorial or Peer reviewed
REVIEWED
Written at
EPFL
EPFL units
Available on Infoscience
November 8, 2017
Use this identifier to reference this record