Skip to main content
triangle with all three sides equal
  • Login
Research
Publications
AI Playground
News
Research Areas
Team
  • People
  • About
  • AI Labs
    • Applied Research
    • 3D DL
    • PAR DL
Collaborations
  • Academic
  • AI Research Residency
  • Government
  • Graduate Fellowships
Careers
  • Internships
  • Research Scientists
  • Publications
  • AI Playground
  • News
  • Research Areas
  • Team
    • People
    • About
    • AI Labs
  • Collaborations
    • Academic
    • AI Research Residency
    • Government
    • Graduate Fellowships
  • Careers
    • Internships
    • Research Scientists

People

 

Donghyuk Lee

Donghyuk Lee is a research scientist in the Architecture Research Group. The primary research focus is enabling high bandwidth and energy efficient memory systems in future GPUs.

 

Research Area(s): 
Circuits and VLSI Design
High Performance Computing
Resilience and Safety
Main Field of Interest: 
Computer Architecture
Google Scholar: 
https://scholar.google.com/citations

Publications

Near-Memory Data Transformation for Efficient Sparse Matrix Multi-Vector Multiplication
DeLTA: GPU Performance Model for Deep Learning Applications with In-depth Memory System Traffic Analysis
What Your DRAM Power Models Aren’t Telling You: Lessons from a Detailed Experimental Study
Reducing Data Transfer Energy by Exploiting Similarity within a Data Transaction
Fine-Grained DRAM: Energy-Efficient DRAM for Extreme Bandwidth Systems
Understanding Reduced-Voltage Operation in Modern DRAM Devices: Experimental Characterization, Analysis, and Mechanisms
Design-Induced Latency Variation in Modern DRAM Chips: Characterization, Analysis, and Latency Reduction Mechanisms
Architecting an Energy-Efficient DRAM System for GPUs
Copyright © 2021 NVIDIA Corporation |  Login |  Legal Information |  Privacy Policy