Pengzhan Hao

Position:         Research Assistant in MOSLab
Education:      Doctoral Student of SUNY at Binghamton
Email:             phao3 at binghamton dot edu

Education

• 2017.08 -                 Ph.D. in Computer Science, Binghamton University, NY, USA
• 2015.08 - 2017.05   M.S in Computer Science, Binghamton University, NY, USA
• 2011.08 - 2015.06   B.E in Software Engineering, Beihang University, Beijing, China

Experience

• 2016.10 -   Research Assistant @ Mobile and Operating System Lab, Binghamton University
  • Advised by Professor Yifan Zhang
  • Edge-based distributed deep learning
    • Implemented a system for time-sensitive, edge-based distributed neural network training scenarios with state-of-the-art performance.
    • Applied empirical experiments to show potential opportunities in Edge-based/Mobile-based distributed neural network training.
    • Investigated heterogeneous-aware training tasks scheduling problem.
  • Edge Computing
    • Envision edge computing’s work-flow in end device performance(latency, traffic) improving.
    • Applied empirical experiments on different edge computing work-flow and propose/implement related solutions.
    • Design and implement an edge-based multi-tier system for decreasing bandwidth usage from endpoint mobile device.
• 2013.10 - 2014.09   Research Intern @ Advanced computer application technology engineering Research Center, Ministry of Education, China
  • Analyzed real-time routing problem and reliable algorithm design
  • Led and developed a routing system mostly backend server program
  • Participated in vehicle internet research
  • Supervised by Juhua Pu, Zhang Xiong, Related project: Urban Computing
• 2013.10 - 2014.01   Research Intern @ Institute of Software, Chinese Academy of Sciences, China
  • Analyzed mobile OS kernel and Localized portion with C language
  • Supervised by Cangzhou Yuan

Research projects

• EDDL: Edge-based distributed deep learning
• Edge Courier: Edge-involved Cloud file synchronization service      Original repository(private)
• Real time routing planner      Original repository(private)
  • Map preprocessing driver

Publications

• EDDL: A Distributed Deep Learning System for Resource-limited Edge Computing Environment.
  Pengzhan Hao, and Yifan Zhang The Sixth ACM/IEEE Symposium on Edge Computing (ACM/IEEE SEC)
  San Jose, CA, USA, 2021.

• A Case for Web Service Bandwidth Reduction on Mobile Devices with Edge-hosted Personal Services
  Yongshu Bai, Pengzhan Hao, and Yifan Zhang The 37th IEEE International Conference on Computer Communications (IEEE INFOCOM)
  Honolulu, HI, USA, 2018.

• EdgeCourier: An Edge-hosted Personal Service for Low-bandwidth Document Synchronization in Mobile Cloud Storage Services [slide]
  Pengzhan Hao, Yongshu Bai, Xin Zhang, and Yifan Zhang
  The 2nd ACM/IEEE Symposium on Edge Computing (ACM/IEEE SEC)
  San Jose, CA, USA, 2017.

• Poster: Securing Device Inputs for Smartphones Using Hypervisor Based Approach
  Xin Zhang, Yongshu Bai, Pengzhan Hao, and Yifan Zhang
  The 15th ACM International Conference on Mobile Systems, Applications, and Services (ACM MobiSys)
  Niagara Falls, NY, USA, 2017.

• Poster: EPS - Edge-hosted Personal Services for Mobile Users [poster]
  Pengzhan Hao, Yongshu Bai, Xin Zhang, and Yifan Zhang
  The 15th ACM International Conference on Mobile Systems, Applications, and Services (ACM MobiSys)
  Niagara Falls, NY, USA, 2017.

Teaching

• Operating system (Graduate level)
• Operating system (Undergraduate level)
• Introduction to distributed system (Graduate level)

Misc