Hongyu Miao, Ph.D.

Deep Learning Performance Software Engineer
NVIDIA

miaohongyu12@gmail.com

LinkedIn | Google Scholar

Biography

I'm a deep learning performance software engineer in the TensorRT team at NVIDIA. I received my Ph.D. in Computer Engineering from Purdue University in December 2021. I work on computer systems, including OS, compiler, architecture, and runtime systems for emerging applications and hardware. Recently, I have been focusing on system for machine learning.

Experience

• DL Performance Engineer @NVIDIA (2022 - now)
  TensorRT for high-performance deep learning inference on NVIDIA GPUs.


• Research Asisstant @Purdue University (2015 - 2021)
  System support for machine learning and real-time data analytics by exploiting modern hardware, e.g., many-core CPUs, hybrid memory (DRAM + 3D-stacked memory), and tiny microcontrollers, aiming to improve system performance and enable new use cases.


• Research Intern @Microsoft Research (2017 - 2017)
  System support for stream processing on high-bandwidth memory.


• Research Assistant @ICT-CAS (2012 - 2015)
  OS kernel (memory subsystem & drivers), system virtualization (QEMU & KVM), and RPC systems (with RDMA).

Publications

• Systems Support for Data Analytics by Exploiting Modern Hardware
  Hongyu Miao
  PhD Dissertation, Purdue University
  West Lafayette, IN, December 2021
  [ PDF | Slides ]
  
  
• Towards Out-of-core Neural Networks on Microcontrollers
  Hongyu Miao and Felix Xiaozhu Lin
  The 7th ACM/IEEE Symposium on Edge Computing (SEC 2022)
  Seattle, WA, December 2022
  [ PDF ] Best Paper Award
  
  
• StreamBox-HBM: Stream Analytics on High Bandwidth Hybrid Memory
  Hongyu Miao, Myeongjae Jeon, Gennady Pekhimenko, Kathryn S. McKinley, and Felix Xiaozhu Lin
  The 24th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2019)
  Providence, RI, April 2019
  [ PDF | Slides | Poster | Website ]
  
  
• StreamBox: Modern Stream Processing on a Multicore Machine
  Hongyu Miao, Heejin Park, Myeongjae Jeon, Gennady Pekhimenko, Kathryn S. McKinley, and Felix Xiaozhu Lin
  The 2017 USENIX Annual Technical Conference (USENIX ATC 2017)
  Santa Clara, CA, July 2017
  [ PDF | Slides | Poster | Website ]
  
  
• Tell Your Graphics Stack That the Display Is Circular
  Hongyu Miao and Felix Xiaozhu Lin
  The 17th International Workshop on Mobile Computing Systems and Applications (HotMobile 2016)
  St. Augustine, FL, February 2016
  [ PDF | Slides | Poster ]

Awards

• Best Paper Award, ACM/IEEE SEC 2022


• Student Travel Grants, ASPLOS 2019, ASPLOS 2018, USENIX ATC 2017, and HotMobile 2016


• Purdue ECE Fellowship 2015