CAFEEN: A Cooperative Approach for Energy Efficient NoCs with Multi-Agent Reinforcement Learning

TL;DR

This paper introduces CAFEEN, a multi-agent reinforcement learning framework that dynamically manages power-gating in NoC architectures, significantly reducing energy consumption while maintaining performance.

Contribution

The paper presents a novel hybrid power-gating approach combining heuristic and reinforcement learning techniques for energy-efficient NoCs.

Findings

Reduces energy by 2.60x for single workloads

Achieves 4.37x energy savings for multi-application workloads

Balances power efficiency with performance adaptively

Abstract

In emerging high-performance Network-on-Chip (NoC) architectures, efficient power management is crucial to minimize energy consumption. We propose a novel framework called CAFEEN that employs both heuristic-based fine-grained and machine learning-based coarse-grained power-gating for energy-efficient NoCs. CAFEEN uses a fine-grained method to activate only essential NoC buffers during lower network loads. It switches to a coarse-grained method at peak loads to minimize compounding wake-up overhead using multi-agent reinforcement learning. Results show that CAFEEN adaptively balances power-efficiency with performance, reducing total energy by 2.60x for single application workloads and 4.37x for multi-application workloads, compared to state-of-the-art NoC power-gating frameworks.