CAP-RAM: A Charge-Domain In-Memory Computing 6T-SRAM for Accurate and Precision-Programmable CNN Inference
نویسندگان
چکیده
A compact, accurate, and bitwidth-programmable in-memory computing (IMC) static random-access memory (SRAM) macro, named CAP-RAM, is presented for energy-efficient convolutional neural network (CNN) inference. It leverages a novel charge-domain multiply-and-accumulate (MAC) mechanism circuitry to achieve superior linearity under process variations compared conventional IMC designs. The adopted semi-parallel architecture efficiently stores filters from multiple CNN layers by sharing eight standard 6T SRAM cells with one MAC circuit. Moreover, up six levels of bit-width weights two encoding schemes input activations are supported. 7-bit charge-injection SAR (ciSAR) analog-to-digital converter (ADC) getting rid sample hold (S&H) input/reference buffers further improves the overall energy efficiency throughput. 65-nm prototype validates excellent accuracy CAP-RAM. single 512x128 macro complete pruned quantized model 98.8% inference on MNIST data set 89.0% CIFAR-10 set, 573.4-giga operations per second (GOPS) peak throughput 49.4-tera (TOPS)/W efficiency.
منابع مشابه
A 481pJ/decision 3.4M decision/s Multifunctional Deep In-memory Inference Processor using Standard 6T SRAM Array
This paper describes a multi-functional deep in-memory processor for inference applications. Deep inmemory processing is achieved by embedding pitch-matched low-SNR analog processing into a standard 6T 16KB SRAM array in 65 nm CMOS. Four applications are demonstrated. The prototype achieves up to 5.6X (9.7X estimated for multi-bank scenario) energy savings with negligible (≤1%) accuracy degrada...
متن کاملComparison of Conventional 6T SRAM cell and FinFET based 6T SRAM Cell Parameters at 45nm Technology
When working for low power application the main estimation is to reduce leakage components and parameters. This stanza explores a vast link towards low leakage power SRAM cells using new technology and devices. The RAM contains bi-stable cross coupled latch which has V_th higher in write mode access MOSFET (Metal Oxide Semiconductor Field Effect Transistor) and lower V_th in read access mode MO...
متن کاملpassivity in waiting for godot and endgame: a psychoanalytic reading
this study intends to investigate samuel beckett’s waiting for godot and endgame under the lacanian psychoanalysis. it begins by explaining the most important concepts of lacanian psychoanalysis. the beckettian characters are studied regarding their state of unconscious, and not the state of consciousness as is common in most beckett studies. according to lacan, language plays the sole role in ...
Memory Access Optimization and RAM Inference for Pipeline Vectorization
This paper describes memory access optimization in the context of pipeline vectorization, a method for synthesizing hardware pipelines in recon gurable systems from software program loops. Since many algorithms for recon gurable coprocessors are I/O bound, the throughput of the coprocessor is determined by the external memory accesses. Thus access optimizations directly improve the system's per...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Journal of Solid-state Circuits
سال: 2021
ISSN: ['0018-9200', '1558-173X']
DOI: https://doi.org/10.1109/jssc.2021.3056447