11 papers
TileSight: A First-Principles Tile-Centric Analytical GPU Performance Model from Cores to Clusters
Zhiwen Mo, Yu Cheng, Lei Wang +12
Recent GPU programming frameworks such as Triton, TileLang, and CUDA Tile adopt tiles as first-class primitives, making tile-centric programming the prevailing approach for high-pe…
Exploring the Efficiency of 3D-Stacked AI Chip Architecture for LLM Inference with Voxel
Yiqi Liu, Noelle Crawford, Michael Wang +2
To overcome the well-known memory bottleneck of AI chips, 3D stacked architectures that employ advanced packaging technology with high-density through-silicon vias (TSVs) pins have…
DeepStack: Facilitating Co-Design Exploration of 3D DRAM-Stacked Accelerators for Distributed LLM Inference
Zhiwen Mo, Guoyu Li, Hao Mark Chen +11
Advances in hybrid bonding and packaging have driven growing interest in 3D DRAM-stacked AI accelerators. As large language models (LLMs) scale to hundreds of billions or trillions…
MoE-CAP: Benchmarking Cost, Accuracy and Performance of Sparse Mixture-of-Experts Systems
Yinsicheng Jiang, Yao Fu, Yeqi Huang +13
The sparse Mixture-of-Experts (MoE) architecture is increasingly favored for scaling Large Language Models (LLMs) efficiently, but it depends on heterogeneous compute and memory re…
ELK: Exploring the Efficiency of Inter-core Connected AI Chips with Deep Learning Compiler Techniques
Yiqi Liu, Yuqi Xue, Noelle Crawford +2
To meet the increasing demand of deep learning (DL) models, AI chips are employing both off-chip memory (e.g., HBM) and high-bandwidth low-latency interconnect for direct inter-cor…
LUT Tensor Core: A Software-Hardware Co-Design for LUT-Based Low-Bit LLM Inference
Zhiwen Mo, Lei Wang, Jianyu Wei +8
Large Language Model (LLM) inference becomes resource-intensive, prompting a shift toward low-bit model weights to reduce the memory footprint and improve efficiency. Such low-bit…