5 papers
Energy-Efficient LLM Serving via Disaggregated Attention--FFN and Flexible Frequency Scaling
Cunchen Hu, Liangliang Xu, Tian Liu +9
Large language model (LLM) serving spans diverse applications with stringent service-level objectives (SLOs), often requiring GPUs to run at maximum frequencies and increasing ener…
STQuant: Spatio-Temporal Adaptive Framework for Optimizer Quantization in Large Multimodal Model Training
Minglu Liu, Cunchen Hu, Liangliang Xu +3
Quantization is an effective way to reduce the memory cost of large-scale model training. However, most existing methods adopt fixed-precision policies, which ignore the fact that…
A Systematic Characterization of LLM Inference on GPUs
Haonan Wang, Xuxin Xiao, Mingyu Yan +8
This work presents a systematic characterization of Large Language Model (LLM) inference to address fragmented understanding. Through comprehensive experiments, we establish a four…
DDiT: Dynamic Resource Allocation for Diffusion Transformer Model Serving
Heyang Huang, Cunchen Hu, Jiaqi Zhu +7
The Text-to-Video (T2V) model aims to generate dynamic and expressive videos from textual prompts. The generation pipeline typically involves multiple modules, such as language enc…
MemServe: Context Caching for Disaggregated LLM Serving with Elastic Memory Pool
Cunchen Hu, Heyang Huang, Junhao Hu +8
Large language model (LLM) serving has transformed from stateless to stateful systems, utilizing techniques like context caching and disaggregated inference. These optimizations ex…