GraphPerf-RT: Graph-Driven Performance Modeling with Calibrated Uncertainty for OpenMP Scheduling on Heterogeneous Embedded SoCs

Pivezhandi, Mohammad; Banisharif, Mahdi; Bakhshan, Saeed; Saifullah, Abusayeed; Jannesari, Ali

doi:10.24963/ijcai.2026/722

GraphPerf-RT: Graph-Driven Performance Modeling with Calibrated Uncertainty for OpenMP Scheduling on Heterogeneous Embedded SoCs

Mohammad Pivezhandi, Mahdi Banisharif, Saeed Bakhshan, Abusayeed Saifullah, Ali Jannesari

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. AI4Tech: AI Enabling Technologies. Pages 6491-6499.

https://doi.org/10.24963/ijcai.2026/722

Keywords

Advanced AI4Tech: AI4Tech foundations
Advanced AI4Tech: Deep AI4Tech
Advanced AI4Tech: Data-driven AI4Tech

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, an AI technology achieving deep learning accuracy at heuristic speeds (2-7ms). GraphPerf-RT is, to our knowledge, the first graph-grounded infrastructure unifying 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. The architecture supports multi-task evidential heads with Normal-Inverse-Gamma 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.

Mohammad Pivezhandi, Mahdi Banisharif, Saeed Bakhshan, Abusayeed Saifullah, and Ali Jannesari. 2026. GraphPerf-RT: Graph-Driven Performance Modeling with Calibrated Uncertainty for OpenMP Scheduling on Heterogeneous Embedded SoCs. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 6491–6499. https://doi.org/10.24963/ijcai.2026/722

@inproceedings{ijcai2026p722,
  title     = {GraphPerf-RT: Graph-Driven Performance Modeling with Calibrated Uncertainty for OpenMP Scheduling on Heterogeneous Embedded SoCs},
  author    = {Pivezhandi, Mohammad and Banisharif, Mahdi and Bakhshan, Saeed and Saifullah, Abusayeed and Jannesari, Ali},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {6491--6499},
  year      = {2026},
  month     = {8},
  note      = {AI4Tech: AI Enabling Technologies},
  doi       = {10.24963/ijcai.2026/722},
  url       = {https://doi.org/10.24963/ijcai.2026/722},
}

TY  - CPAPER
TI  - GraphPerf-RT: Graph-Driven Performance Modeling with Calibrated Uncertainty for OpenMP Scheduling on Heterogeneous Embedded SoCs
AU  - Pivezhandi, Mohammad
AU  - Banisharif, Mahdi
AU  - Bakhshan, Saeed
AU  - Saifullah, Abusayeed
AU  - Jannesari, Ali
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 6491
EP  - 6499
DO  - 10.24963/ijcai.2026/722
UR  - https://doi.org/10.24963/ijcai.2026/722
ER  -