Students with Interests

If you are good at two or more of the following skills, we shall definitely meet and talk. (If you are a undergraduate, one is sufficient)
  • Decent programming skills (C, C++, Python, Rust, etc.)
  • Data Management Programming or Kernel Optimization (SQL, CQL, NewSQL, PostgreSQL, Cassandra, Redis, Zookeeper, Docker, etc.)
  • Hardware hands-on experience (Verilog/VHDL/Chisel, HLS C/C++/OpenCL, CUDA, OpenCL)
  • Hardware simulators and modeling (gem5, gpgpu-sim, Multi2Sim, etc.)
  • Modeling and machine learning (Matlab, PyTorch, TF, Keros, PaddlePaddle)
For more information, please feel free to contact me using the email address on my page.

On-going Projects

  • Energy and Battery-aware Federated Learning
    We are trying to build a sustainable learning service that considers system performance metrics such as energy, battery use, etc., in a scale-out manner. For that, we build up cloud-edge infrastructure to sustain long-term and correct key-value data store among places.
  • A Holistic Consistency Balancing across All Replicas
    We are trying to extend the original raft into the practical scenarios. That is providing scalable and cheap distributed services within the Raft protocol. The main contribution of this research is to extending the scope of a strong consensus algorithm into a very unreliable platform and make it work statistically in practice.
  • Database Optimization
    We are interested in in-memory processing for enabling efficient personal intelligent services (PIS). We believe the storage management of the knowledge is the bottleneck of the PIS, and in-memory process shall be the way to better the performance of PIS on personal smart devices. As such, we are working on providing a materialized view for Graphic database, which is used to represent the inner-connection in the knowledge base. In this case, we are able to provide PIS in the remote device without offloading the computation into the central servers.
  • Data Mining and Learning for SMILES in Chemistry
    We are developing a novel and fast learning method to speedup the simulation of chemistry synthesis, such computation in the wild is our prime concern on design a better computing system for such typical streaming data services.
  • A Practical Black Water Raft Library
    Based on ETCD-Raft and implemented by C++, eRaft is a high-performance C++ Raft library. This project is mainly developed by graduates from our GOOD lab. The Raft algorithm shall be accredited to Dr. Diego Ongaro. At present, our project has been included in the official distribution. We hope to explore the possibility of optimizing the existing algorithms on the basis of realizing a stable practical Raft library. If you are interested, please join us. Anyone interested may refer project.