Tianyu Guo (郭天宇)

E-mail : guoty9[at]mail2.sysu.edu.cn

---

About

I’m a second-year PH.D. student of Computer Science and Technology at Sun Yat-Sen University advised by Assoc. Prof. Xianwei Zhang. I completed bachelor degree at Xidian University. My reasearch insterest lies in GPU architecture, HPC and LLM inference. Check out my resume for more details.

Publications

• Tianyu Guo, Xuanteng Huang, Kan Wu, Xianwei Zhang and Nong Xiao, SMILE: LLC-based Shared Memory Expansion to Improve GPU Thread Level Parallelism, The 61st ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, United States, June 2024.

Experience

• Research (LLM Inference) intern at Tencent [2024]

• Teaching Assistant of “SYSU-DCS3013 : Computer Architecture” [2022f]

  release SYSU-ARCH LAB

Projects

• Optimize GEMM step by step

• PTX-EMU: A simple emulator for CUDA program

• A simple CNN training framework support on CPU and GPU(CUDNN)

• GEMM by CUDA WMMA (tensor core)

• Mouse Controler(APP): Use your phone as a mouse

• eChat(APP): A simple Internet chatroom

• BCI: Web interface for mindwave MOBILE2(NeuroSky)

Presentations & HW & Dissertation

• DAC’24 SMILE “SMILE: LLC-based Shared Memory Expansion to Improve GPU Thread Level Parallelism”

• Weekly Sharing CrossKV “CrossKV: Reuse KV Cache Across Requests”

• Weekly Paper Sharing SC23 “Frontier: Exploring Exascale”

• Weekly Paper Sharing MLSYS23 “AUTOSCRATCH: ML-OPTIMIZED CACHE MANAGEMENT FOR INFERENCE-ORIENTED GPUS”

• Work report 22-23

• Weekly Paper Sharing HPCA23 “DIMM-Link: Enabling Efficient Inter-DIMM Communication for Near-Memory Processing”

• AI final Homework “A Convolutional Neural Network Framework support on CPU and GPU”

• Bachelor’s dissertation “General Computing optimization for GPU based on Cache management”