Kuan-Chieh Hsu (Jay)
Hi, I am a PhD candidate in the department of Computer Science and Engineering at University of California, Riverside. I am a research assistant supervised by Dr. Hung-Wei Tseng in Extreme Storage & Computer Architecture Laboratory(ESCAL). I received my Master degree in Computer and Communication Engineering from National Cheng Kung University, Taiwan, in 2016. There, I worked on a HSA-Compatible General-purpose GPU simulator project in CASLAB and supervised by Dr. Chung-Ho Chen. Prior to this, I received my Bachelor degree with CGPA 3.91/4.00 in Electrical Engineering from National Cheng Kung University in 2014. Also, I was an exchange student in Electrical Engineering and Computer Science department at Auburn University in Fall 2013.
[Curriculum Vitae] [Google Scholar] [Linkedin] [firstname.lastname@example.org] [personal website]
- Kuan-Chieh Hsu, Hung-Wei Tseng, “GPTPU: Accelerating Applications using Edge Tensor Processing Units”, SC ’21: International Conference for High Performance Computing, Networking, Storage and Analysis, 2021 [arXiv] [github]
- Kuan-Chieh Hsu, Chung-Ho Chen, “Performance Prediction Model on HSA-Compatible General-Purpose GPU System,” Thesis for Master of Science, Institute of Computer and Communication Engineering, National Cheng Kung University, July, 2016
- Yun-Chi Huang, Kuan-Chieh Hsu, Wan-Shan Hsieh, Chen-Chieh Wang, Chia-Han Lu, Chung-Ho Chen, “Dynamic SIMD Re- convergence with Paired-Path Comparison, ” in the 2016 IEEE International Symposium on Circuit and System (ISCAS), 2016
- Hen-Yi Chen, Chung-Ho Chen, Yun-Chi Huang, Kuan-Chieh Hsu, Chen-Chieh Wang, “An HSAIL conformed GPU platform, ” in the International Conference on Applied System Innovation, ICASI 2015
- Heterogeneous Edge General-Purpose Computing Framework
- Developing a programming framework that allows machine learning workloads to leverage both EdgeTPUs and GPU cores on an ARM-based embedded system platform using C++.
- GPTPU: Accelerating Applications using Edge Tensor Processing Units
- Developed a programming framework and the system library for Google’s EdgeTPU AI/ML accelerators to achieve 2.46x speedup and 40% energy saving compared to high-performance CPU cores for a set of matrix applications using C++.
- Redesigned the algorithm and implemented the library function for general matrix multiplications using 2D convolutions to more efficiently use the EdgeTPU hardware and achieve 41.2x speedup using C++.
- Autonomous Near-Data Processing on Intelligent SSD
- Implemented Financial and linear algebra applications in Python with C-wrapper Python functions to invoke GPU kernels. The implementation offers our proposed Python runtime to achieve the same level of performance of C runtime.
- Technological Development of ESL Full System for Heterogeneous Computing Platform
- funded by Ministry of Science and Technology, R.O.C. Sep. 2014 – Jun. 2016
- This sub-project focuses on architecture and software framework construction.
- The main goal of the corresponding principal project is to develop a Heterogeneous System Architecture (HSA) platform including the key system technology and applications. In this multi-stage fast-prototyping work I developed scalable memory subsystem for GPU streaming-multiprocessors including hierarchical cache system, NoC module, and memory controllers.
- Implementation of HSA-compatible GPU Architecture
- Funded by ITRI, Taiwan Jan. 2015 – Dec. 2015
- Developing early-staged heterogeneous system simulation for OpenCL and OpenGL applications using hardware-style abstraction.
- Modularized Course Development Program
- Funded by Ministry of Education, R.O.C. Jun. 2015 – Aug. 2015
- Designed and implemented ARM assembly courses for Computer Organization practice course based on a full-system CPU simulator. This work was accepted as standard material by more than ten universities for undergraduate computer organization course in Taiwan.
- Heterogeneous Edge General-Purpose Computing Framework
University of California, Riverside 2019 – Present
- PhD candidate in department of Computer Science and Engineering
North Carolina State University 2018 – 2019 (transferred out)
- PhD student in department of Computer Science
National Cheng Kung University 2014 – 2016
- M.S. in Institute of Computer and Communication Engineering, GPA: 4.00/4.00
National Cheng Kung University 2010 – 2014
- B.S. in department of Electrical Engineering, GPA: 3.91/4.00
- Work experience
- Performance team intern, Cerebras System, Sunnyvale, summer 2022
- System architect intern, Samsung Semiconductor, San Jose, summer 2019
- Research assistant, ESCAL, UCR, 2018 – present
- Research assistant, DIRL lab, Academia Sinica, 2017
- K12 STEM educator, Mijo Tech, Taiwan, 2016
- Research assistant, CASLAB, NCKU, 2014 – 2016
- Summer intern, Nuvoton technology corp., 2014
- Course CS 161 A01: Design and Architecture of Computer Systems, summer 2021
- Teaching Assistant
- Course E216401: Logic System and affiliated experiment, 2015
- Course E221700: Computer Organization and affiliated experiment, 2014
- Student volunteer for 2018 IEEE International Symposium on Workload Characterization, Raleigh, NC, Sep. 30 – Oct. 2, 2018
- 4th Taiwan Artificial Intelligence/Data Science Conference coordinator in Academia Sinica, Taiwan, Nov. 9-12, 2017.
- Exchange Student in Department of Computer Science and Software Engineering, Auburn University, AL, USA Aug. 2013 – Dec. 2013
- University Graduate Scholarship from NCSU, 2018 – 2019
- M.S. Distinguished Entrance Scholarship from NCKU, 2014
- Guo-Qing Pan Scholarship, 2013
- Guo-Fu Chen Scholarship, 2013
- Leadtrend Scholarship, 2011
- B.S. Distinguished Entrance Scholarship from NCKU, 2010
- First round award of core system master cultivation program, 2014
- The award of outstanding student for the academic achievement, 2012
- Third place of on-campus competition of Calculus, 2011
- Top 1% reward of on-campus competition of General physics, 2011