HPCA 2026
Sat 31 January - Wed 4 February 2026 Sydney, Australia
co-located with HPCA/CGO/PPoPP/CC 2026
Mon 2 Feb 2026 14:30 - 14:50 at Coogee - LLM Inference Serving Systems Chair(s): Jian Li

Multiple Low-Rank Adapters (Multi-LoRA) are gaining popularity for task-specific Large Language Model (LLM) applications. For Multi-LoRA serving, caching hot LoRAs and KV caches in the GPU memory can improve inference performance. However, existing Multi-LoRA inference systems fail to optimize serving performance like Time-To-First-Toke (TTFT), neglecting usage dependencies when caching LoRAs and KV caches. We therefore propose \textbf{ELORA}, a Multi-LoRA caching system to optimize the serving performance. ELORA comprises a \textit{dependency-aware cache manager} and a \textit{performance-driven cache swapper}. The cache manager maintains the usage dependencies between LoRAs and KV caches during inference with a unified caching pool. The cache swapper determines the swap-in or swap-out of LoRAs and KV caches based on a unified cost model, when the GPU memory is idle or busy, respectively. Experimental results show that ELORA reduces the TTFT by 45.7% on average, compared to state-of-the-art works.

Mon 2 Feb

Displayed time zone: Hobart change

14:10 - 15:30
LLM Inference Serving SystemsMain Conference at Coogee
Chair(s): Jian Li Chinese Academy of Meteorological Sciences
14:10
20m
Talk
Towards Resource-Efficient Serverless LLM Inference with SLINFER
Main Conference
Chuhao Xu Shanghai Jiao Tong University, Zijun Li Shanghai Jiao Tong University, Quan Chen Shanghai Jiao Tong University, China, Han Zhao Shanghai Jiao Tong University, Xueyan Tang Nanyang Technological University, Minyi Guo Shanghai Jiao Tong University
14:30
20m
Talk
ELORA: Efficient LoRA and KV Cache Management for Multi-LoRA LLM Serving
Main Conference
Jiuchen Shi Shanghai Jiao Tong University & The Hong Kong Polytechnic University, Hang Zhang Shanghai Jiao Tong University, Yixiao Wang Shanghai Jiao Tong University, Quan Chen Shanghai Jiao Tong University, China, Yizhou Shan Huawei Cloud, Kaihua Fu Hong Kong University of Science and Technology, Wei Wang Hong Kong University of Science and Technology, Minyi Guo Shanghai Jiao Tong University
14:50
20m
Talk
PASCAL: A Phase-Aware Scheduling Algorithm for Serving Reasoning-based Large Language Models
Main Conference
15:10
20m
Talk
The Cost of Dynamic Reasoning: Demystifying AI Agents and Test-Time Scaling from an AI Infrastructure Perspective
Main Conference
Jiin Kim KAIST, Byeongjun Shin KAIST, Jinha Chung KAIST, Minsoo Rhu KAIST