Welcome to the Parallel and Distributed Systems Laboratory (PADSYS Lab). We are part of the Department of Electrical and Computer Engineering at the University of Florida.
Our research focuses on understanding, modeling, designing, and building real-world computer systems. We work on scalable, efficient, and secure systems for modern computing, with an emphasis on parallel and distributed computing, high-performance communication and I/O, and heterogeneous infrastructures spanning CPUs, GPUs, DPUs, and others. Our recent efforts explore next-generation communication libraries, data movement optimization, and system support for large-scale AI and HPC infrastructures.
We are particularly interested in enabling emerging workloads through co-design across systems, algorithms, and architectures, with strong considerations of performance, scalability, and security. We also collaborate closely with domain scientists to address interdisciplinary challenges in areas such as agriculture, neuroscience, and beyond.
Join Us
We are looking forward to working with self-motivated Ph.D., M.S., and undergraduate students with research interests in parallel and distributed computing systems. Ph.D. students will be supported by full scholarships to cover the tuition fees and living expenses.
Please feel free to contact Prof. Xiaoyi Lu if you have interest.
Links
You can find any of our research projects here.
You can find any of our publications here.
You can find any of our developed software here.
Lab Photos
Recent Publications
- Advancing Data Center Workloads with Data Processing Units
- PULSE: Power Usage Monitoring Leveraging Sensing of Electromagnetic Field
- Understanding the Idiosyncrasies of Emerging BlueField DPUs
- SBMGT: Scaling Bayesian Multinomial Group Testing
- HPC-R1: Characterizing R1-like Large Reasoning Models on HPC
- DPAR: High-Performance, Secure, and Scalable Differential Privacy-based AllReduce
- DPU-KV: On the Benefits of DPU Offloading for In-Memory Key-Value Stores at the Edge
- Accelerating Lossy and Lossless Compression on Emerging BlueField DPU Architectures
- NVMe-oPF: Designing Efficient Priority Schemes for NVMe-over-Fabrics with Multi-Tenancy Support
- SBGT: Scaling Bayesian-based Group Testing for Disease Surveillance
- A Study of Database Performance Sensitivity to Experiment Settings
- NVMe-oAF: Towards Adaptive NVMe-oF for IO-Intensive Workloads on HPC Cloud
- HatRPC: Hint-Accelerated Thrift RPC over RDMA
- DStore: A Fast, Tailless, and Quiescent-Free Object Store for PMEM
- Understanding the Idiosyncrasies of Real Persistent Memory
- RDMP-KV: Designing Remote Direct Memory Persistence Based Key-Value Stores with PMEM
- INEC: Fast and Coherent in-Network Erasure Coding
- TriEC: Tripartite Graph Based Erasure Coding NIC Offload
- UMR-EC: A Unified and Multi-Rail Erasure Coding Library for High-Performance Distributed Storage Systems.