Energy-Efficient Near-Threshold Parallel Computing: The PULPv2 Cluster
| Auteurs | Rossi D., Pullini A., Loi I., Gautschi M., Gurkaynak F.K., Teman A., Constantin J., Burg A., Miro-Panades I., Beigne E., Clermidy F., Flatresse P., Benini L. |
| Year | 2017-0421 |
| Source-Title | IEEE Micro |
| Affiliations | University of Bologna, Italy, ETH Zurich, Switzerland, Univ. of Bologna, Italy, EPFL, Italy, CEA-LETI, France, STMicroelectronics, France, University of Bologna and ETH Zurich, Switzerland |
| 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. © 1981-2012 IEEE. |
| Author-Keywords | body biasing, energy efficiency, parallel processing, power management, UTBB FD-SOI |
| Index-Keywords | Cluster computing, Internet of things, Power management, Programmable logic controllers, System-on-chip, Body biasing, Energy efficient, Internet of Things (IOT), Near thresholds, Parallel computing platform, Parallel processing, Sensor processing, UTBB FD-SOI, Energy efficiency |
| ISSN | 2721732 |
| Lien vers article | Link |