Junmo Lee

Junmo Lee

Auto-generated profile, based on publicly available information

Junmo Lee

School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 January 2024 in Nature Reviews Electrical Engineering
Verified
Journal Article
Closed access
PDF file
ALAnni Lu
Anni Lu
JLJunmo Lee
Junmo Lee
TKTae-Hyeon Kim
Tae-Hyeon Kim
MKMuhammed Ahosan Ul Karim
Muhammed Ahosan Ul Karim
RPRebecca Sejung Park
Rebecca Sejung Park
SYShimeng Yu
Shimeng Yu

21 February 2024 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
JLJunmo Lee
Junmo Lee
ALAnni Lu
Anni Lu
WLWantong Li
Wantong Li
SYShimeng Yu
Shimeng Yu
Over the past decade, numerous compute-in-memory (CIM) platforms have been proposed in the literature. While emerging non-volatile memory based analog CIM (ACIM) has been widely studied, its silicon demonstrations are in the mature legacy node (22 nm or above). As an alternative, digital CIM (DCIM) based on static random access memory (SRAM) is recently drawing significant attention, as it enjoys the scaling benefits with the logic process to the leading-edge node (5 nm or below), and does not suffer from the accuracy loss due to process/voltage/temperature (PVT) variations. To assess the potential of DCIM in the future, we release NeuroSim V1.4, a CIM benchmark framework, which supports advanced technology nodes down to 1 nm node. We project the technology parameters (standard cell, transistor and interconnect) using TCAD device simulations, interconnect modeling, and the available industry/IRDS roadmaps. State-of-the-art technology trends such as fin-depopulation, buried power rail, stacked nanosheet, etc are captured in the updated parameters. Technology scaling down to 1 nm enables DCIM to achieve 1.4 $\sim$ 1.8 $\times$ and 44.1 $\sim$ 63.1 $\times$ higher system-level figure of merit than state-of-the-art 7 nm SRAM-based ACIM and 22 nm RRAM-based ACIM, respectively, for representative workloads such as ResNet18 and ResNet34 inference.

24 October 2025 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
FWFaaiq Waqar
Faaiq Waqar
JKJungyoun Kwak
Jungyoun Kwak
JLJunmo Lee
Junmo Lee
OPOmkar Phadke
Omkar Phadke
MSMinji Shon
Minji Shon
SYShimeng Yu
Shimeng Yu
The Last Level Cache (LLC) is the processor’s critical bridge between on-chip and off-chip memory levels - optimized for high density, high bandwidth, and low operation energy. To date, high-density (HD) SRAM has been the conventional device of choice; however, with the slowing of transistor scaling, as reflected in the industry’s almost identical HD SRAM cell size from 5 nm to 3 nm, alternative solutions such as 3D stacking with advanced packaging (i.e., hybrid bonding) are pursued (as demonstrated in AMD’s V-cache). Escalating data demands necessitate ultra-large on-chip caches to decrease costly off-chip memory movement, pushing the exploration of device technology towards monolithic 3D (M3D) integration, where transistors can be stacked in the back-end-of-line (BEOL) at the interconnect level. M3D integration requires fabrication techniques compatible with a low thermal budget (<400°C). Among promising BEOL device candidates are amorphous oxide semiconductor (AOS) transistors, particularly desirable for their ultra-low leakage (seconds) when used in a gain-cell configuration. This paper examines device, circuit, and system-level tradeoffs made when optimizing BEOL-compatible AOS-based 2-transistor gain cells (2T-GC) for LLC. A cache early-exploration tool, NS-Cache, is developed to model caches in advanced 7 & 3 nm nodes and is integrated with the Gem5 simulator to systematically benchmark the impact of the newfound density/performance when compared to HD-SRAM, MRAM, and 1T1C eDRAM alternatives for LLC.

21 March 2025 in IEEE Electron Device Letters
Verified
Journal Article
Closed access
OPOmkar Phadke
Omkar Phadke
PHPo-Kai Hsu
Po-Kai Hsu
YLYuan-Chun Luo
Yuan-Chun Luo
JSJanak Sharda
Janak Sharda
VGVaidehi Garg
Vaidehi Garg
SYShimeng Yu
Shimeng Yu
In this letter, a ferroelectric field effect transistor (FeFET)-based charge-domain ternary content addressable memory (TCAM) is experimentally demonstrated for in-memory search. Specifically, foundry based FeFETs are configured in non-volatile capacitor (nvCAP) mode for charge-domain computation. A single TCAM cell consisting of 2 nvCAPs (i.e. two FeFETs) is used to perform the Store: bit ‘0’, ‘1’, ‘X’ and Search: bit ‘0’, ‘1’, ‘X’ operations. Further, SPICE simulations are performed to evaluate the feasibility of a scaled nvCAP device in a large TCAM array. The low CON/COFF ratio of a scaled device can lead to false-positive matchings, which can be countered by introducing a dummy column and input dependent charge cancellation. Finally, the charge-domain nvCAP TCAM is benchmarked against other proposed candidates, showing a ~1.5× improvement in energy efficiency over FeFET based current domain TCAM, and a ~3.3× improvement compared to resistive random access memory (RRAM) based TCAM.

25 October 2024 in IEEE Transactions on Electron Devices
Verified
Journal Article
Closed access
SDSunbin Deng
Sunbin Deng
JKJungyoun Kwak
Jungyoun Kwak
JLJunmo Lee
Junmo Lee
DCDyutimoy Chakraborty
Dyutimoy Chakraborty
JSJaewon Shin
Jaewon Shin
SDSuman Datta
Suman Datta
This work reports the first experimental demonstration of on-chip switched-capacitor (SC) dc–dc voltage converters, where two types of back-end-of-line (BEOL) compatible device components—amorphous oxide semiconductor (AOS) power transistors and high-voltage (HV) superlattice MIM capacitors—were monolithically integrated for 12–6 V conversion with an efficiency of 82.8%. To implement more efficient dual- $\textit{V}_{\text{th}}$ circuit design, both enhancement-and depletion-mode AOS power transistors were developed. Oxygen anneal was performed to enhance transistor reliability, achieving up to a 6.5 $\times$ reduction in threshold voltage shift. For the MIM capacitors, with a dielectric stack composed of Hf $_{\text{0.25}}$ Zr $_{\text{0.75}}$ O $_{\text{2}}$ superlattice laminates and Al $_{\text{2}}$ O $_{\text{3}}$ interlayers, the projected lifetime could exceed 10 years at 120 $^{\circ}$ C and 6 V bias, ensuring long-term reliable operation of the converters. Machine learning (ML)-assisted device modeling and circuit simulation results indicated that 4-interleaved design and 3-D capacitor design could further optimize the converter with the decrease of voltage ripple by $\sim$ 60% and the increase of power density to $>$ 1 W/mm $^{\text{2}}$ . The developed on-chip SC dc–dc converters are a competitive option for implementing efficient vertical power delivery networks (PDNs) in heterogeneous 3-D integrated circuits (ICs).

23 June 2024
Verified
Conference Paper
Open access
ALAnni Lu
Anni Lu
JLJunmo Lee
Junmo Lee
YLYuan-Chun Luo
Yuan-Chun Luo
HLHai Li
Hai Li
IYIan Young
Ian Young
SYShimeng Yu
Shimeng Yu
Combinatorial optimization problems (COP) are NP-hard and intractable to solve using conventional computing. The Ising model-based annealer has gained increasing attention recently due to its efficiency and speed in finding approximate solutions. However, Ising solvers for travelling salesman problems (TSP) usually suffer from a scalability issue due to quadratically increasing number of spins. In this paper, we propose a digital computing-in-memory (CIM) based clustered annealer to solve tens of thousands of city-scale TSP with only a few mega-byte (MB) of static random access memory (SRAM), using hierarchical clustering to solve input sparsity and digital CIM flexibility to solve weight sparsity. The intrinsic process variations between SRAM devices are utilized to generate the noisy bit errors during pseudo-read under reduced supply voltage, realizing the annealing process. The design space of cluster size and programmability is explored to understand the trade-offs of solution quality and hardware cost, for TSP scale ranging from 3080 to 85900 cities. The proposed design speeds up the convergence by >109× with 1013× improvement on functionally normalized area and power.