2025 Pre-OSDI/ATC Workshop

Northeastern Systems Research Group is organizing a half-day pre-OSDI/ATC workshop on the Northeastern campus, featuring OSDI/ATC practice talks and informal discussions to foster community and build connections.

Schedule (subject to change)

Time Speaker Title
1:00 PM - 1:10 PMOpening remarks
1:10 PM - 2:10 PMSystems for AI Training and Inference I
Yuxuan Jiang (University of Michigan)Training with Confidence: Catching Silent Errors in Deep Learning Training with Automated Proactive Checks
Jinkun Lin (New York University)Understanding Stragglers in Large Model Training Using What-if Analysis
Jiali Wang (Shanghai Jiao Tong University)Colocating ML Inference and Training with Fast GPU Memory Handover
2:10 PM - 2:40 PMBreak
2:40 PM - 3:40 PMSystems for AI Training and Inference II
Wenxin Zheng (Shanghai Jiao Tong University)SAVE: Software-Implemented Fault Tolerance for Model Inference against GPU Memory Bit Flips
Congjie He (University of Edinburgh)WaferLLM: Large Language Model Inference at Wafer Scale
Ruofan Wu (University of Michigan)PluS: Highly Efficient and Expandable ML Compiler with Pluggable Graph Schedules
3:40 PM - 4:10 PMBreak
4:10 PM - 5:20 PMBridging Boundaries in System Designs
Evangelos (Vagos) Lamprou (Brown University)The Koala Benchmarks for the Shell
Leon Schuermann (Princeton University)Building Bridges: Safe Interactions with Foreign Languages through Omniglot
Hakim Weatherspoon (Cornell University)Towards a Practical, Scalable Oblivious Reconfigurable Network
Yu Hua (Huazhong University of Science and Technology)Memory as a Fabric: How Networking Reshapes Big Memory Systems
5:20 PM - 5:30 PMClosing remarks

Contact

Please send any questions to Ji-Yong Shin <j.shin@northeastern.edu> and Cheng Tan <c.tan@northeastern.edu>.

Acknowledgements

This event is sponsered by Systems Research Group at Northeastern University.