Conglong Li

Conglong Li is a fifth-year Ph.D. student in Computer Science Department at Carnegie Mellon University advised by Professor David G. Andersen. He received both B.S. and M.S. in Computer Science from Rice University in 2013 and 2014, respectively. During his M.S. study, he was advised by Professor Alan L. Cox and supported by the Graduate Research Fellowship.

In general, Conglong works on improving performance and resource efficiency of various computer systems with a focus on algorithms and data structures. In particular, he is focusing on the intersection of systems and machine learning:

• Systems for ML: building efficient systems for ML applications.
• ML for Systems: leveraging ML techniques to build intelligent systems.

Curriculum vitae

Email: conglong dot li @gmail.com

Publication

• Google Scholar
• Scaling Video Analytics on Constrained Edge Nodes.
  Christopher Canel, Thomas Kim, Giulio Zhou, Conglong Li, Hyeontaek Lim, David G. Andersen, Michael Kaminsky, Subramanya R. Dulloor.
  In Proc. SysML 2019.
• Better Caching in Search Advertising Systems with Rapid Refresh Predictions.
  Conglong Li, David G. Andersen, Qiang Fu, Sameh Elnikety, Yuxiong He.
  In Proc. WWW 2018.
• Workload Analysis and Caching Strategies for Search Advertising Systems.
  Conglong Li, David G. Andersen, Qiang Fu, Sameh Elnikety, Yuxiong He.
  In Proc. ACM SoCC 2017.
• Using Indirect Routing to Recover from Network Traffic Scheduling Estimation Error.
  Conglong Li, Matthew K. Mukerjee, David G. Andersen, Srinivasan Seshan, Michael Kaminsky, George Porter, Alex C. Snoeren.
  In Proc. ACM/IEEE ANCS 2017.
• Scheduling Techniques for Hybrid Circuit/Packet Networks.
  He Liu, Matthew K. Mukerjee, Conglong Li, Nicolas Feltman, George Papen, Stefan Savage, Srinivasan Seshan, Geoffrey M. Voelker, David G. Andersen, Michael Kaminsky, George Porter, Alex C. Snoeren.
  In Proc. ACM CoNEXT 2015.
• GD-Wheel: A Cost-Aware Replacement Policy for Key-Value Stores.
  Conglong Li, Alan L. Cox.
  In Proc. ACM EuroSys 2015.
• Reducing DRAM Row Activations with Eager Read/Write Clustering.
  Myeongjae Jeon, Conglong Li, Alan L. Cox, Scott Rixner.
  In ACM TACO 2013.
• GD-Wheel: A Cost-Aware Replacement Policy for Key-Value Stores.
  Conglong Li, Alan L. Cox.
  In 7th Workshop on Large-Scale Distributed Systems and Middleware (LADIS 2013).

Internship

• Summer 2019 Research Intern at Microsoft AI & Research.
  Mentor: Yuxiong He.
  Worked on improving approximate nearest neighbor search performance (this project started in 2018). Designed ML models (GBDT) to predict the search termination condition for each query. Evaluations demonstrate up to 10.8 times speedup under the same accuracy targets. Submitted a paper to a top tier conference. Working on publishing the code and shipping it in Microsoft products.
• Summer 2017 Software Engineer Intern at Microsoft AI & Research.
  Mentor: Ramu Movva, Yuxiong He.
  Worked on designing caching strategies for Bing Ads. Designed ML models (GBDT) to provide intelligent cache refresh decisions. Simulations on production traces demonstrate a potential 35.2 to 106.1 million dollars net profit gain in a quarter. Transferred the project to developing team to ship it in product. Project ended up with a paper published at WWW 2018.
• Summer 2016 Research Intern at Microsoft Research.
  Mentor: Yuxiong He, Ramu Movva.
  Worked on designing caching strategies for Bing Ads. Designed domain-specific caching heuristics to save ads selection cost and improve net profit. Simulations on production traces demonstrate a potential 20.7 to 70.5 million dollars net profit gain in a quarter. Project ended up with a paper published at SoCC 2017.