PATRONoC: Parallel AXI Transport Reducing Overhead for Networks-on-Chip targeting Multi-Accelerator DNN Platforms at the Edge

TL;DR

PATRONoC is an open-source, AXI-compliant NoC fabric designed for multi-core DNN platforms, significantly reducing overhead and increasing throughput and area efficiency compared to classical NoCs.

Contribution

It introduces PATRONoC, a novel parallel AXI-based NoC architecture optimized for DNN workloads, with improved efficiency and throughput over existing solutions.

Findings

34% higher area efficiency at 1 GHz

2-8X throughput improvement over baseline

up to 350 GiB/s throughput on workloads

Abstract

Emerging deep neural network (DNN) applications require high-performance multi-core hardware acceleration with large data bursts. Classical network-on-chips (NoCs) use serial packet-based protocols suffering from significant protocol translation overheads towards the endpoints. This paper proposes PATRONoC, an open-source fully AXI-compliant NoC fabric to better address the specific needs of multi-core DNN computing platforms. Evaluation of PATRONoC in a 2D-mesh topology shows 34% higher area efficiency compared to a state-of-the-art classical NoC at 1 GHz. PATRONoC's throughput outperforms a baseline NoC by 2-8X on uniform random traffic and provides a high aggregated throughput of up to 350 GiB/s on synthetic and DNN workload traffic.