Zejia Lin

According to our database, Zejia Lin authored at least 6 papers between 2022 and 2025.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of four.

2025

Bullet: Boosting GPU Utilization for LLM Serving via Dynamic Spatial-Temporal Orchestration.

Zejia Lin, Hongxin Xu, Guanyi Chen, Xianwei Zhang, Yutong Lu

CoRR, April, 2025

Mpache: Interaction Aware Multi-level Cache Bypassing on GPUs.

Mengyue Xi, Tianyu Guo, Xuanteng Huang, Zejia Lin, Xianwei Zhang

Proceedings of the 30th Asia and South Pacific Design Automation Conference, 2025

2024

MixPert: Optimizing Mixed-Precision Floating-Point Emulation on GPU Integer Tensor Cores.

Zejia Lin, Aoyuan Sun, Xianwei Zhang, Yutong Lu

Proceedings of the 25th ACM SIGPLAN/SIGBED International Conference on Languages, 2024

2023

Hay: Enhancing GPU Sharing Performance With Two-Level Scheduling for Ray.

Lianghong Huang, Zejia Lin, Wei Liu, Xianwei Zhang

Proceedings of the 29th IEEE International Conference on Parallel and Distributed Systems, 2023

KeSCo: Compiler-based Kernel Scheduling for Multi-task GPU Applications.

Zejia Lin, Zewei Mo, Xuanteng Huang, Xianwei Zhang, Yutong Lu

Proceedings of the 41st IEEE International Conference on Computer Design, 2023

2022

moTuner: a compiler-based auto-tuning approach for mixed-precision operators.

Zewei Mo, Zejia Lin, Xianwei Zhang, Yutong Lu

Proceedings of the CF '22: 19th ACM International Conference on Computing Frontiers, Turin, Italy, May 17, 2022