News

01/2018 Start COMP 421 TA

11/2017 Attend SC'17

10/2017 Refactor DeepPerf Solver (DeepPerf)

09/2017 Take part in Openmp Tools Interface

08/2017 Enroll in Rice

07/2017 Take a vacation

06/2017 Attend ICS'17

05/2017 Refactor gSpan (gBolt)

04/2017 Update homepage

03/2017 Keep contributing to torch

11/2016 Release blitz v0.1: blitz v0.1

10/2016 Update blog: Programming KNL--Architecture

About Me

I am a graduate student in Rice University, advised by Professor John Mellor-Crummey. Previously, I studied at Institute of Computing Technology, Chinese Academy of Sciences in Professor Guangming Tan's PAA group. Prior to that, I was an undergraduate student in Yunnan University, advised by Professor Wei Zhou.

Brief CV

Education

Ph.D. Computer Science, 2017 - 2023 (expected)

School of Engineering, Rice University

M.S. Computer Science, 2014 - 2017

Institute of Computing Technology, Chinese Academy of Sciences

B.E. Network Engineering, 2010 - 2014

School of Software, Yunnan University

Experience

Research Intern, Apr.2017 - Aug.2017

Nvidia Inc, Beijing

Research Assistant, Jun.2015 - Aug.2017

Nvidia-Sugon-ICT Deep Learning Joint Laboratory, Institute of Computing Technology

Research Assistant, Jan.2013 - July.2014

Intelligent Web Laboratory, School of Software, Yunnan University

SDE Intern, Oct.2013 - Feb.2014

Baidu Inc, Beijing

Research

My major research areas are parallel systems and concurrent algorithms. I focus on optimizing parallel systems and developing efficient concurrent data structures on modern architectures. I also participate in data mining contests as a hobby.

Projects (Full List)

  • GPU Performance Analysis Tools

    March.2017 - current
    We have proposed a paper about GPU performance analysis. Beyond the previous work, we are going to extend our framework for wider applications, multiple kernels, and several architectures. Besides, designing user-friendly interfaces is also a primary goal.

Publications (Full List)

Talks

  • Deep Learning on Modern Architectures
    April.2017, Institute of Computing Technology, Chinese Academy of Sciences
    Discussed how state-of-the-art deep learning libraries optimize computations by utilizing architectural features.
  • Convolution Methods
    July.2016, Institute of Computing Technology, Chinese Academy of Sciences
    Introduced various kinds of convolution methods and analyzed their complexities, memory consumptions, and data access patterns.