USENIX ATC '25 - DEEPSERVE: Serverless Large Language Model Serving at Scale

DEEPSERVE: Serverless Large Language Model Serving at Scale

Junhao Hu, Peking University and Key Lab of HCST (PKU), MOE; Jiang Xu, Zhixia Liu, Yulong He, Yuetao Chen, Hao Xu, Jiang Liu, Jie Meng, Baoquan Zhang, Shining Wan, Gengyuan Dan, Zhiyu Dong, Zhihao Ren, and Changhong Liu, Huawei Cloud; Tao Xie, Key Lab of HCST (PKU), MOE and Peking University; Dayun Lin, Qin Zhang, Yue Yu, Hao Feng, Xusheng Chen, and Yizhou Shan, Huawei Cloud

In this paper, we propose DEEPSERVE, a scalable and serverless AI platform designed to efficiently serve large language models (LLMs) at scale in cloud environments. DEEPSERVE addresses key challenges such as resource allocation, serving efficiency, and cold start latencies through four main design components. First, DEEPSERVE uses a simple serverless abstraction called the request-job-task model, which helps manage diverse AI workloads across post-training and model-serving tasks.
Second, DEEPSERVE integrates an in-house serving engine named FLOWSERVE using a microkernel-inspired design, NPU-centric execution, and SPMD-based parallelism to optimize LLM serving.
Third, DEEPSERVE includes novel scheduling policies tailored for a configuration with both PD-disaggregated and PD-colocated instances. Fourth, DEEPSERVE includes optimizations such as pre-warmed pods, DRAM pre-loading, and NPU-fork, which allow DEEPSERVE to scale up to 64 instances in seconds. DEEPSERVE has been in production for over a year, operating on a large Ascend NPU cluster and providing industry-standard APIs for fine-tuning, agent serving, and model serving to our customers.

View the full USENIX ATC '25 program at https://www.usenix.org/conference/atc...