GreenDyGNN: Runtime-Adaptive Energy-Efficient Communication for Distributed GNN Training

TL;DR

GreenDyGNN introduces a runtime-adaptive caching strategy using reinforcement learning to significantly reduce energy consumption in distributed GNN training under network congestion.

Contribution

It formulates cache management as a sequential decision problem and employs a Double-DQN agent for adaptive cache rebuilding during training.

Findings

GreenDyGNN reduces energy consumption by up to 43% under congestion.

It closely matches optimal performance in clean network conditions.

Adaptive cache management outperforms static policies significantly.

Abstract

Distributed GNN training is dominated by remote feature fetching, which can be very costly. Multi-hop neighborhood sampling crosses partition boundaries and triggers fine-grained RPCs whose fixed initiation cost and GPU-stall latency waste energy. Prior systems try to reduce this overhead with presampling and static caching, but cache policies cannot react to runtime network variation. We show that under time-varying congestion, static caching can increase energy by up to 45% because a fixed rebuild schedule is insufficient. We present GreenDyGNN, which formulates cache window management as a sequential decision problem. GreenDyGNN performs intra-epoch cache rebuilds and uses a Double-DQN agent, trained in a calibrated simulator with domain-randomized congestion, to adapt rebuild window size and per-owner cache allocation at each boundary. An asynchronous double-buffered pipeline makes adaptation effectively free. Under congestion, GreenDyGNN cuts total energy by up to 43% over Default DGL and 4-24% over the best static policy, while closely matching the optimum under clean conditions.