Skip to main navigation Skip to search Skip to main content

An SRAM-Based Hybrid Computation-in-Memory Macro Using Current-Reused Differential CCO

  • Injun Choi
  • , Edward Jongyoon Choi
  • , Donghyeon Yi
  • , Yoontae Jung
  • , Hoyong Seong
  • , Hyuntak Jeon
  • , Soon Jae Kweon
  • , Ik Joon Chang
  • , Sohmyung Ha
  • , Minkyu Je
  • Korea Advanced Institute of Science and Technology
  • Korean Agency for Defense Development
  • Kyung Hee University
  • New York University Abu Dhabi
  • New York University

Research output: Contribution to journalArticlepeer-review

12 Scopus citations

Abstract

This work presents a 4 kb 8T-SRAM computation-in-memory (CIM) macro based on hybrid computation using digital in-memory-array computing (DIMAC) and phase-domain near-memory-array computing (PNMAC). By employing multiple local dual-column arrays (LDCAs), bit-wise multiplications are computed digitally in memory with high energy efficiency and throughput. The PNMAC performs the summation and accumulation in parallel with a high dynamic range by using a proposed steering-DAC-based differential current-controlled-oscillator (DCCO). After the phase-domain accumulation is completed, only a one-time digital conversion needs to be performed using a phase quantizer with negligible phase-to-digital conversion overhead. Moreover, by effectively reusing the steered current to accumulate the multiplication results fed from the DIMAC, the power consumption of the PNMAC can be greatly reduced. The macro fabricated in a 65 nm process achieves 22.4TOPS/W peak energy efficiency and 19.03~mu text{W} power consumption with a 59.8% zero-skipping rate, which is 96.05times lower than state of the art.

Original languageEnglish
Pages (from-to)536-546
Number of pages11
JournalIEEE Journal on Emerging and Selected Topics in Circuits and Systems
Volume12
Issue number2
DOIs
StatePublished - 1 Jun 2022

Bibliographical note

Publisher Copyright:
© 2011 IEEE.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 7 - Affordable and Clean Energy
    SDG 7 Affordable and Clean Energy

Keywords

  • Convolutional neural network (CNN)
  • SRAM
  • computation in memory (CIM)
  • differential current-controlled-oscillator (DCCO)
  • digital in-memory-array computing (DIMAC)
  • phase-domain near-memory-array computing (PNMAC)

Fingerprint

Dive into the research topics of 'An SRAM-Based Hybrid Computation-in-Memory Macro Using Current-Reused Differential CCO'. Together they form a unique fingerprint.

Cite this