Liquid Silicon: A Nonvolatile Fully Programmable Processing-in-Memory Processor With Monolithically Integrated ReRAM
Authors:
- 13 January 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 55 (4) , 908-919
Abstract
The slowdown of the CMOS technology scaling, and the trade-off between efficiency and flexibility have fueled the exploration into novel architectures with emerging postCMOS technology [e.g., resistive-RAM (RRAM)]. In this article, a nonvolatile fully programmable processing-in-memory (PIM) processor named liquid silicon is demonstrated, which combines the superior programmability of general-purpose computing devices [e.g., field-programmable gate array (FPGA)] and the high efficiency of domain-specific accelerators. Besides the general computing applications, liquid silicon is particularly well suited for artificial intelligence (AI)/machine learning and big data applications, which not only poses high computational/memory demand but also evolves rapidly. To fabricate the liquid silicon chip, the HfO₂ RRAM is monolithically integrated on top of the commercial 130 nm CMOS. Our measurement confirms that liquid silicon chip can operate reliably at a low voltage of 650 mV. It achieves 60.9 TOPS/W in performing neural network (NN) inferences, and 480 GOPS/W in performing content-based similarity search (a key big data application) at a nominal voltage supply of 1.2 V, showing 3x and 100x improvement over the state-of-the-art domain-specific CMOS-/RRAM-based accelerators. In addition, it outperforms the latest nonvolatile FPGA in energy efficiency by 3x in general computing applications.Keywords
Funding Information
- Defense Advanced Research Projects Agency (D16AP00122)
This publication has 11 references indexed in Scilit:
- A 141 UW, 2.46 PJ/Neuron Binarized Convolutional Neural Network Based Self-Learning Speech Recognition Processor in 28NM CMOS2018
- BinarEye: An always-on energy-accuracy-scalable binary CNN processor with all memory on chip in 28nm CMOS2018
- Liquid Silicon-Monona2018
- BRein memory: A 13-layer 4.2 K neuron/0.8 M synapse binary/ternary reconfigurable in-memory deep neural network accelerator in 65 nm CMOS2017
- Fundamental variability limits of filament-based RRAM2016
- CAPI: A Coherent Accelerator Processor InterfaceIBM Journal of Research and Development, 2015
- DRAMA: An Architecture for Accelerated Processing Near MemoryIEEE Computer Architecture Letters, 2014
- 1 Mb 0.41 µm² 2T-2R Cell Nonvolatile TCAM With Two-Bit Encoding and Clocked Self-Referenced SensingIEEE Journal of Solid-State Circuits, 2013
- Nonvolatile 3D-FPGA with monolithically stacked RRAM-based configuration memory2012
- Similarity search and locality sensitive hashing using ternary content addressable memories2010