Problem
Research questions and friction points this paper is trying to address.
Goodput Maximization
Large Language Model
Edge Inference
Service Level Objective
Wireless Edge Networks
Goodput Maximization for Large Language Model Edge Inference: A Two-Phase Maskable PPO Approach
🤖 AI Summary
本文提出了一种两阶段可掩码近端策略优化算法，通过优化任务卸载决策和带宽分配来提高无线边缘网络中大型语言模型推理服务的系统好吞吐量。
📝 Abstract
This paper presents a novel two-phase maskable proximal policy optimization (TP-MPPO) algorithm, which maximizes the system goodput counting request throughput with strict service level objective (SLO) compliance for large language model (LLM) inference services in wireless edge networks. In the first phase of TP-MPPO, we optimize the task offloading decisions by MPPO with action masking mechanism, effectively avoiding exploring invalid actions and reducing the action space. In the second phase, closed-form solutions are derived for uplink bandwidth allocation; a greedy algorithm is designed for downlink bandwidth allocation to provide immediate rewards for the MPPO in the next round. The two stages alternate till convergence. Simulation results demonstrate that TP-MPPO can improve the system reward by 33.3%--87.5% compared to its benchmarks and achieve the highest goodput.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Two-Phase Maskable PPO
Task Offloading Decisions
Action Masking Mechanism
Uplink Bandwidth Allocation
Downlink Bandwidth Allocation
🔎 Similar Papers
No similar papers found.
