Jim Dai

Yueying Li, Jim Dai, Tianyi Peng

As demand for Large Language Models (LLMs) and AI agents rapidly grows, optimizing systems for efficient LLM inference becomes critical. While significant efforts have focused on system-level engineering, little is explored from a mathematical modeling and queuing perspective. In this paper, we aim to develop the queuing fundamentals for large language model (LLM) inference, bridging the gap between the queueing theory and LLM system communities. In particular, we study the throughput aspect in LLM inference systems. We prove that a large class of 'work-conserving' scheduling algorithms can achieve maximum throughput for individual inference LLM engine, highlighting 'work-conserving' as a key design principle in practice. In a network of LLM agents, work-conserving scheduling alone is insufficient, particularly when facing specific workload structures and multi-class workflows that require more sophisticated scheduling strategies. Evaluations of real-world systems show that Orca and Sarathi-serve are throughput-optimal, reassuring practitioners, while FasterTransformer and vanilla vLLM are not maximally stable and should be used with caution. Our results highlight the substantial benefits that the queueing community can offer in improving LLM inference systems and call for more interdisciplinary development.

Jim Dai, Manxi Wu, Zhanhao Zhang

Pioneering companies such as Waymo have deployed robo-taxi services in several U.S. cities. These robo-taxis are electric vehicles, and their operations require the joint optimization of ride matching, vehicle repositioning, and charging scheduling in a stochastic environment. We model the operations of the ride-hailing system with robo-taxis as a discrete-time, average-reward Markov Decision Process with an infinite horizon. As the fleet size grows, dispatching becomes challenging, as both the system state space and the fleet dispatching action space grow exponentially with the number of vehicles. To address this, we introduce a scalable deep reinforcement learning algorithm, called Atomic Proximal Policy Optimization (Atomic-PPO), that reduces the action space using atomic action decomposition. We evaluate our algorithm using real-world NYC for-hire vehicle trip records and measure its performance by the long-run average reward achieved by the dispatching policy, relative to a fluid-based upper bound. Our experiments demonstrate the superior performance of Atomic-PPO compared to benchmark methods. Furthermore, we conduct extensive numerical experiments to analyze the efficient allocation of charging facilities and assess the impact of vehicle range and charger speed on system performance.