Image of Tsung Tai Yeh
EC 707, Engineering Building 3, 1001 University Road,
Hsinchu 300, Taiwan
886-03-5712121#54727
ttyeh@cs.nctu.edu.tw
LAB: EC 120

News

  • Nov. 2020 Congrats Tsung Tai, paper @HPCA 2021
  • Oct. 2020 Congrats Meng-Chien et al. won MakeNTU --InnOvaTion VIA Prize

    • Welcome

    I am an assistant professor in the department of computer science at National Chiao Tung University. My research is cross the bounary among computer architecture, computer systems, and programming languages.

    I am always looking for undergraduate and graduate students who are excited about computer architecture, systems, and programming languages research. Please email me with a copy of your resume or CV if you are interested in working with me.

    Research

    My research aims to design high throughput, low latency, and high energy efficiency domain-specific accelerators and their systems.


    My research domains:
    · Computer Architecture
    · Computer Systems
    · Memory and Storage Systems
    · Domain-specific Accelerators (GPU, Neural Processing Units)

    Bio

    I received a Ph.D in Electrical and Computer Engineering, Purdue University in 2020 and was advised by Timothy G. Rogers. I obtained my M.S. in Institute of Information Systems and Applications, National Tsing Hua University, Taiwan. Previously, I worked at AMD Research, Purdue University, and Academia Sinica. My research work was nominated by the Best Paper Award (PPoPP 2017). I was also the recipient of the Lynn Fellowship at Purdue University. To know more about me, please visit my CV.

    Selected Publications

  • Deadline-Aware Offloading for High-Throughput Accelerators, Tsung Tai Yeh, Matthew D. Sinclair, Brad Beckmann, Timothy G Rogers, in the 27th IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2021. Acceptance rate: (63/258) = 24.4%
  • Dimensionality-Aware Redundant SIMT Instruction Elimination, Tsung Tai Yeh, Roland Green, Timothy G. Rogers, in ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 2020.Acceptance rate: 86/476 = 18%
  • Pagoda: Fine-Grained GPU Resource Virtualization for Narrow Tasks, Tsung Tai Yeh, Amit Sabne, Putt Sakdhnagool, Rudolf Eigenmann, Timothy G. Rogers, in Proceedings of ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP), 2017. Acceptance rate: 39/132 = 22%, Best Paper Nominee

    • Grant Funding

  • Optimizing Domain-Specific Accelerator Hardware Resource Utilization, Tsung Tai Yeh, 109-2222-E-009-009-MY2, MOST, 11/01/2020 - 10/31/2022

    Invited Talk

  • When Designs of Computer Architecture Meet Machine Learning, NCTU CS, Freshman Seminar, 11/11/2020, [slide]
  • When Microprocessor Designs Meet Machine Learning, NCTU CS, Junior Seminar, 11/16/2020, [slide]
  • Scaling Performance and Energy Efficiency Through Domain-Specific Accelerators, NTHU/NCU CS, Graduate Student Seminar, 11/18/2020, 11/25,2020, [slide]
  • When GPU Architecture Designs Meet Machine Learning, NTHU CS, Advanced Computer Architecture Class, 1/7/2021, [slide]
    • counter free