PowerFlow-DNN: Compiler-Directed Fine-Grained Power Orchestration for End-to-End Edge AI Inference

TL;DR

PowerFlow-DNN is a compiler-driven framework that optimizes energy efficiency in edge AI accelerators by intelligently scheduling power states across layers within real-time constraints, significantly reducing energy consumption.

Contribution

It introduces a novel end-to-end power scheduling method that accounts for inter-layer effects and practical hardware constraints, enabling near-optimal energy savings in ultra-low-power edge AI systems.

Findings

Achieves energy within 0.68% of the optimal ILP solution.

Reduces energy consumption by up to 37% over baseline.

Operates efficiently over a massive schedule space with 10^{160} possibilities.

Abstract

Edge AI systems often operate under stringent energy and volume constraints that demand extreme efficiency under limited battery capacity, with requirements worsening as intelligent capability demands advance. Prior literature suggests that fine-grained power orchestration, including DVFS and power gating, enables significant energy efficiency benefits that cannot be left unexploited, while still exhibiting unexplored challenges. We observe that layer-level approaches incur unintended overheads due to inter-layer coupling of power control decisions, and that jointly managing these mechanisms under practical constraints such as limited voltage rails and transition overheads leads to a rapidly growing combinatorial schedule space. To address this, we propose PowerFlow-DNN, a compiler-directed framework for end-to-end power-state orchestration in ultra-low-power accelerators. By constructing a rigorous problem formulation for deadline-constrained, real-time, periodic inference as a unified inter-layer power-scheduling problem, our framework enables automated discovery of energy-minimal power-state schedules that adhere to a deadline while accounting for end-to-end, inter-layer impacts. We evaluate the framework on a DNN accelerator VLSI implementation in TSMC 40nm technology. Across representative edge networks, we show that PowerFlow-DNN discovers near-optimal solutions under the discretized formulation and achieves energy within 0.68\% of the exact ILP oracle, reducing energy by up to 37\% compared to an aggressive baseline without power orchestration, while reasoning over a combinatorial schedule space of over $10^{160}$ possible power-state assignments, yet operating on a structured layered state graph that enables efficient optimization, achieving up to 2.14$\times$ solver speedup via lightweight pruning.