Hi, I’m Jialong Liu, an M.S. student in Software Engineering at Shanghai Jiao Tong University, advised by Prof. Rong Chen in the Institute of Parallel and Distributed Systems (IPADS). I expect to graduate in March 2028.

My interests lie at the intersection of systems and AI infrastructure.I enjoy turning systems ideas into working software and evaluating them under realistic resource constraints. I am always happy to discuss systems research, open-source collaboration, and engineering opportunities.

Research Interests

• Operating systems and runtime systems
• Scheduling for GPUs, NPUs, and heterogeneous accelerators
• AI inference systems and model serving
• Memory management and resource multiplexing
• High-performance and latency-sensitive systems

Papers

• Weihang Shen, Mingcong Han, Jialong Liu, Rong Chen, and Haibo Chen. “XSched: Preemptive Scheduling for Diverse XPUs.” Proceedings of the USENIX Annual Technical Conference (OSDI), 2025.

Projects

vLLM Switch

vLLM Switch enables faster model switching in vLLM by extending sleep mode with memory- and disk-backed model-weight snapshots. The vLLM Switch Controller provides an external control plane for request-driven switching and co-location across long-lived vLLM backends.

XSched

XSched is a scheduling framework that brings flexible preemptive scheduling to diverse accelerators, including GPUs, NPUs, ASICs, and FPGAs. My contributions include implementing Intel Level Zero support for Intel iGPUs and NPUs, porting XSched to Windows, and building its cross-platform IPC library.

XServer

XServer is a dynamic memory scheduler on top of a memory-adjustable inference engine, enabling concurrent AI workloads to operate under tight memory budgets. This project explores memory-aware resource multiplexing for heterogeneous AI platforms.

OhDiff

OhDiff is a high-performance differential backup system. I led its early systems design and designed both simulation-based and real-machine evaluations.

Teaching

From September 2025 to June 2026, I served as a teaching assistant for Introduction to Computer Systems. My responsibilities included assisting with course delivery, participating in the design and implementation of Agent Lab, and providing student support for topics related to systems and agents.

Awards

• 2026.07 Excellent Teaching Assistant Award, Shanghai Jiao Tong University
• 2025.06 Outstanding Graduate Student Award, Shanghai Jiao Tong University
• 2024.10 First Prize, China Graduate Operating System Open Source Innovation Competition