Yufeng Gu

I am a Research Scientist at AI System Co-Design, Meta Platform, Inc. I obtained my Ph.D. at University of Michigan, advised by Prof. Reetuparna Das. I worked closely with Prof. David Blaauw and Prof. Satish Narayanasamy. My research focuses on computer architecture and system, and has been selected as Communications of ACM Research Highlights. Before UMich, I obtained a bachelor's degree from Zhejiang University in 2020.

Email  /  CV  /  Linkedin  /  Google Scholar  /  Blogs

Research

My research focuses on computer architecture, hardware/software co-design, processing-in-memory, resource provisioning and scheduling. I built novel hardware and system solutions on accelerating large scale emerging applications, including generative artificial intelligence (GenAI) and precision health:

• Data movement-centric architecture:
  • Processing-in-memory system (ASPLOS'25, HPCA'25)
  • Data access acceleration (ISCA'25)
  • GPU Fabric Scheduling (APSys'26)
• Scientific computing and genome sequencing analysis:
  • Workload characterization (IISWC'25, ISPASS'21)
  • Domain-specific accelerators (CACM'25, ISCA'23)
  • Hardware security (DAC'26)

News

Selected Publications

Unlocking Software-defined GPU Fabric Scheduling in the LLM Era
Danyang Chen*, Yufeng Gu*, Yibo Huang, Chengxuan Pei, Peichun Hua, Yang Zhou, Yunming Xiao
APSys 2026

GenDP: A Framework of Dynamic Programming Acceleration for Genome Sequencing Analysis (Invited Paper)
Yufeng Gu, Arun Subramaniyan, Tim Dunn, Alireza Khadem, Kuan-yu Chen, Somnath Paul, Md Vasimuddin, Sanchit Misra, David Blaauw, Satish Narayanasamy, Reetuparna Das
CACM 2025  /  Paper  /  Technical Perspective  /  University News

DX100: A Programmable Data Access Accelerator for Indirection
Alireza Khadem, Kamalavasan Kamalakkannan, Zhenyan Zhu, Akash Poptani, Yufeng Gu, Jered Benjamin Dominguez-Trujillo, Nishil Talati, Daichi Fujiki, Scott Mahlke, Galen Shipman, Reetuparna Das
ISCA 2025  /  Paper  /  Code     

PIM Is All You Need: A CXL-Enabled GPU-Free System for Large Language Model Inference
Yufeng Gu*, Alireza Khadem*, Sumanth Umesh, Ning Liang, Xavier Servot, Onur Mutlu, Ravi Iyer, and Reetuparna Das
ASPLOS 2025  /  Paper  /  Slides  /  Code     

Multi-Dimensional Vector ISA Extension for Mobile In-Cache Computing
Alireza Khadem, Daichi Fujiki, Hilbert Chen, Yufeng Gu, Nishil Talati, Scott Mahlke, Reetuparna Das
HPCA 2025  /  Paper  /  Code     
Distinguished Artifact Honorable Mention

GenDP: A Framework of Dynamic Programming Acceleration for Genome Sequencing Analysis
Yufeng Gu, Arun Subramaniyan, Tim Dunn, Alireza Khadem, Kuan-yu Chen, Somnath Paul, Md Vasimuddin, Sanchit Misra, David Blaauw, Satish Narayanasamy, Reetuparna Das
ISCA 2023  /  Paper  /  Slides  /  Code  /  Lightning Talk     
Communications of ACM Research Highlights  /  Technical Perspective

Awards and Honors

• Outstanding Young Speaker Award at APPT 2025.
  (Hardware-Software Co-Design for Scalable LLM Inference)
• Distinguished Artifact Honorable Mention in HPCA 2025, selected from 3/29 artifacts.
  (Multi-Dimensional Vector ISA Extension for Mobile In-Cache Computing)
• Communications of ACM Research Highlights, selected among 24 papers from all ACM conferences in 2023.
  (GenDP: A Framework of Dynamic Programming Acceleration for Genome Sequencing Analysis)
• Rackham Graduate Student Research Grant at University of Michigan.
  (Pangenome Sequence Alignment Benchmark Suite, Role: PI, $3,000)
• Rackham Conference Travel Grant at University of Michigan, 2023, 2025.
• Student Travel Grant for ISCA 2023, HPCA 2025.
• Summer@EPFL Fellowship (2% applicants awarded), 2019.
• Tang Lixin Fellowship (60/60,000+ students awarded), 2017, 2018, 2019.

Talks

• [12/2025] Architecture and System Co-Design for Scalable LLM Inference at Chinese University of Hong Kong (CUHK).
• [07/2025] Hardware-Software Co-Design for Scalable LLM Inference PhD Forum at APPT 2025.
• [06/2025] CXL-enabled PIM system for LLM inference at the SAFARI Live Seminar.
• [03/2025] CXL-enabled PIM system for LLM inference at the MCCSys workshop co-located with ASPLOS 2025.
• [03/2024] Genomics Benchmark Suite and Accelerator Design on the Computer Architecture Seminar at UCF.
• [02/2024] Genomics Benchmark Suite and Accelerator Design at Cornel University.
• [12/2023] Genomics Benchmark Suite and Accelerator Design on the Peisu Xia Forum at ICT, CAS.

Teaching

• Guest Lecture in EECS 570: Parallel Computer Architecture at University of Michigan.
  "CXL-enabled PIM system for scalable LLM inference", Fall 2025 and Spring 2026.
• Instructor, Programming module at the Discovery Engineering Program, University of Michigan.

Services

• Program Committee
• HPCA 2027
• Journal reviewer
  • IEEE Transactions on Computers (TC)
  • IEEE Transactions on Big Data (TBD)
  • IEEE Computer Architecture Letters (CAL)
  • Journal of Systems Architecture (JSA)
• Artifact evaluation reviewer for IISWC 2021, ISCA 2023, ISCA 2024, ISCA 2025.
• University of Michigan graduate program admission committee, 2021, 2022.