Title
HERMES Core – A 14nm CMOS and PCM-based In-Memory Compute Core using an array of 300ps/LSB Linearized CCO-based ADCs and local digital processing
Abstract
We present a 256×256 in-memory compute (IMC) core designed and fabricated in 14nm CMOS with backend-integrated multi-level phase-change memory (PCM). It comprises 256 linearized current controlled oscillator (CCO)-based ADCs at a compact 4µm pitch and a local digital processing unit performing affine scaling and ReLU operations. A novel frequency-linearization technique for CCOs is introduced, leading to accurate on-chip matrix-vector-multiply (MVM) when operating over 1 GHz. Measured classification accuracies on MNIST and CIFAR-10 datasets are presented when two cores are employed for deep learning (DL) inference. The measured energy efficiency is 10.5 TOPS/W at a performance density of 1.59 TOPS/mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> .
Year
DOI
Venue
2021
10.23919/VLSICircuits52068.2021.9492362
2021 Symposium on VLSI Technology
Keywords
DocType
ISSN
In-memory computing,PCM,ADC,deep learning
Conference
0743-1562
ISBN
Citations 
PageRank 
978-1-6654-3009-8
0
0.34
References 
Authors
0
24