GraphPerf-RT: A Graph-Driven Performance Model for Hardware-Aware Scheduling of OpenMP Codes

TL;DR

GraphPerf-RT is a graph neural network model that accurately predicts performance metrics for real-time, risk-aware scheduling of OpenMP codes on embedded platforms, integrating structural and runtime information.

Contribution

It introduces the first unified graph model combining task DAGs, code semantics, and runtime context for hardware-aware scheduling, with calibrated uncertainty estimation.

Findings

Achieves R^2=0.81 on makespan prediction

Enables 66% makespan reduction with RL integration

Maintains zero thermal violations in experiments

Abstract

Autonomous AI agents on embedded platforms require real-time, risk-aware scheduling under resource and thermal constraints. Classical heuristics struggle with workload irregularity, tabular regressors discard structural information, and model-free reinforcement learning (RL) risks overheating. We introduce GraphPerf-RT, a graph neural network surrogate achieving deep learning accuracy at heuristic speeds (2-7ms). GraphPerf-RT is, to our knowledge, the first to unify task DAG topology, CFG-derived code semantics, and runtime context (per-core DVFS, thermal state, utilization) in a heterogeneous graph with typed edges encoding precedence, placement, and contention. Evidential regression with Normal-Inverse-Gamma priors provides calibrated uncertainty; we validate on makespan prediction for risk-aware scheduling. Experiments on three ARM platforms (Jetson TX2, Orin NX, RUBIK Pi) achieve R^2 = 0.81 on log-transformed makespan with Spearman rho = 0.95 and conservative uncertainty calibration (PICP = 99.9% at 95% confidence). Integration with four RL methods demonstrates that multi-agent model-based RL with GraphPerf-RT as the world model achieves 66% makespan reduction and 82% energy reduction versus model-free baselines, with zero thermal violations.