I am a graduate student at University of Toronto, and I am fortunate to be supervised by Natalie Enright Jerger.
Generally, I am interested in solving problems relating to software infrastructure and heterogeneous hardware in data centers, including cloud systems, and machine learning systems.
I worked as a Research Intern in Networking Research Group at Microsoft Research on research projects of resource disaggregation and SmartNIC-enabled systems.
Fiat Lux.
Click here to see my academic geneology, all the way back to the 19th century!
representing it so as to
make the solution transparent.
-- Herbert Simon, Sciences of the Artificial
Research Interests
- Server Architecture
- Machine Learning (Networked) Systems
- Operating Systems
- RPC Systems
- I/O Device Hardware (NIC, SmartNIC, FPGA, DPU, SSD, GPU)
- I/O Virtualization
- CXL
Publications
-
Youmu: Efficient Columnar Data Pipeline for LLM Training
8th Machine Learning and Systems (MLSys), 2025. [pdf] [slide]
-
SmartNIC-enabled Live Migration for Storage Optimized VMs
15th ACM SIGOPS Asia-Pacific Workshop on Systems (APSys), 2024. [pdf] [slide]
-
The Case of Unsustainable CPU Affinity
2nd Workshop on Sustainable Computer Systems Design and Implementation (HotCarbon), 2023. [pdf] [slide]
-
Altocumulus: Scalable Scheduling for Nanosecond-scale Remote Procedure Calls
55th IEEE/ACM International Symposium on Microarchitecture (MICRO), 2022. [pdf] [slide] [blog]
-
What Can Chiplets Bring to Multi-Tenant Clouds?
1st Workshop on Cloud, IEEE/ACM International Symposium on Microarchitecture (MICRO), 2021. [pdf]
[slide]