Ming Liu

I am an Assistant Professor in the Computer Sciences Department at the University of Wisconsin-Madison (since 2021). I’m a member of the NetLab and madNets groups. I study networking and systems, with a focus on building efficient, robust, and practical computing systems for emerging hardware. My research spans two complementary directions. The first, HW4Net, explores how modern hardware communication fabrics, such as load-store interconnects and programmable networks, reshape the design of networked systems. The second, Net4HW, develops networking and system support for new hardware architectures, including chiplet-based SoCs, domain-specific accelerators, and disaggregated storage.

I am actively looking for students. If you are interested, feel free to email me.

Research

Recently, we build applications, systems, protocols, and utilities for rack/cluster-scale infrastructures.

• Scale-(In|Up) Load-Store Interconnect
  • Projects: ChipletNet, SD-LSI
  • cSwitch enhances Linux scheduling via a chiplet-centric switching architecture (SOSP’26).
  • PingPoint dissects communication behaviors of the Accelerator Chiplet Network (SIGCOMM’26).
  • MemChannel develops a CSFQ-inspired transport layer for switched CXL memory pooling (NSDI’26).
  • PathFinder tracks and analyzes CXL.mem using Intel PMUs (SIGCOMM’25).
  • MegaStation realizes the MIMO baseband processing on a single-node supercomputer (NSDI’25).
• Disaggregated Storage
  • Projects: PDS
  • SANarch compares the switched and switchless architecture for disaggregated SAN (NSDI’26).
  • TapDB understands and optimizes database pushdown over storage disaggregation (ASPLOS’26).
  • ntprof is a profiling utility for characterizing and analyzing the NVMe-over-TCP protocol (NSDI’25).
  • Flint builds an elastic block storage over EBOFs using shadow views (NSDI’25).
• Programmable Networks
  • Projects: CleanNIC, dSys-PNF
  • SID realizes software-interposed datapath via Elastic QPs for rack-scale interconnects (SIGCOMM’26).
  • SG-IOV introduces socket-granular I/O virtualization to offload container networks (ASPLOS’26).
  • SCR provides a framework enabling packet-granular software control over the harware transport (NSDI’25).

Students

Zerui Guo, Wentao Hou, Yuyuan Kang, Joontaek Oh

Service

Program Committee Member: NSDI 2026, ASPLOS 2026, APNet 2025, SOSP 2024, HotInfra 2024, APnet 2024, OSDI 2024, NSDI 2024, HotInfra 2023, APnet 2023, SYSTOR 2023, SIGCOMM 2022, SYSTOR 2021