Transport Optimization for Streaming Systems

Follow me on GitHub

Transport Optimization for Streaming Systems

TranSys is designed to achieve high-performance interactive streaming services for high-resolution, low-latency transport. There are three sub-projects:

  • Metis (Homepage). It provides an interpretation method for the deep learning-based networking systems, with a case study on the video streaming system.
  • PiTree (Homepage). It provides a conversion method from heavyweight adaptive bit-rate algorithms into lightweight decision trees.
  • PiTree Dataset (Homepage). It provides a dataset of network conditions that collected by us for the emulation of high-resolution video streaming.


  • Zili Meng, Yaning Guo, Yixin Shen, Jing Chen, Chao Zhou, Minhu Wang, Jia Zhang, Mingwei Xu, Chen Sun, Hongxin Hu. Practically Deploying Heavyweight Adaptive Bitrate Algorithms With Teacher-Student Learning. IEEE/ACM Transactions on Networking (ToN) 2021.
  • Zili Meng, Minhu Wang, Jiasong Bai, Mingwei Xu, Hongzi Mao, Hongxin Hu. Interpreting Deep Learning-Based Networking Systems. ACM SIGCOMM 2020.
  • Zili Meng, Jing Chen, Yaning Guo, Chen Sun, Hongxin Hu, Mingwei Xu. PiTree: Practical Implementation of ABR Algorithms Using Decision Trees. ACM Multimedia 2019.

Contributing Members

(sorted by the first name)


The TranSys project has been supported by the National Key R&D Program of China (2017YFB0801701) and National Science Foundation of China (61625203, 61832013, 61872426). Zili is also supported by a Microsoft Research Asia Fellowship.


For any questions, please send an email to