The End-Nodes of the Internet of Things (IoT) require extreme energy efficiency coupled with wide power performance operating range. Fully-depleted SOI (FD-SOI) is an attractive technology for ultra-low power and wide-range operation as it offers compelling options to tune power, performance, area (PPA) at design time as well as at run time. This paper describes Quentin: an MCU-class (32bit) open source RISC-V SoC featuring an autonomous I/O subsystem optimized to deal with the wide variety of sensors available in IoT end-nodes, coupled with a processor optimized for near threshold computation and a heterogeneous (standard-cell and SRAM) memory architecture to better exploit the low-voltage capabilities of 22nm FDX technology. The system runs up to 2400 million equivalent RV32IMC instructions per second (MOPS) and achieves best power density of 6 μ W/MHz, resulting into an energy efficiency of 433 MOPS/mW.
Quentin: an ultra-low-power PULPissimo SoC in 22nm FDX / Schiavone P.D.; Rossi D.; Pullini A.; Di Mauro A.; Conti F.; Benini L.. - STAMPA. - (2018), pp. 8640145.1-8640145.3. (Intervento presentato al convegno 2018 IEEE SOI-3D-Subthreshold Microelectronics Technology Unified Conference, S3S 2018 tenutosi a Hyatt Regency, San Francisco Airport, usa nel 2018) [10.1109/S3S.2018.8640145].
Quentin: an ultra-low-power PULPissimo SoC in 22nm FDX
Rossi D.;Conti F.;Benini L.
2018
Abstract
The End-Nodes of the Internet of Things (IoT) require extreme energy efficiency coupled with wide power performance operating range. Fully-depleted SOI (FD-SOI) is an attractive technology for ultra-low power and wide-range operation as it offers compelling options to tune power, performance, area (PPA) at design time as well as at run time. This paper describes Quentin: an MCU-class (32bit) open source RISC-V SoC featuring an autonomous I/O subsystem optimized to deal with the wide variety of sensors available in IoT end-nodes, coupled with a processor optimized for near threshold computation and a heterogeneous (standard-cell and SRAM) memory architecture to better exploit the low-voltage capabilities of 22nm FDX technology. The system runs up to 2400 million equivalent RV32IMC instructions per second (MOPS) and achieves best power density of 6 μ W/MHz, resulting into an energy efficiency of 433 MOPS/mW.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
