Dataflow-Architecture Co-Design for 2.5D DNN Accelerators using Wireless Network-on-Package

TL;DR

This paper introduces WIENNA, a wireless NoP-based 2.5D DNN accelerator that significantly improves throughput and energy efficiency by leveraging high-bandwidth multicasting over interposer-based designs.

Contribution

The paper proposes WIENNA, a novel wireless NoP architecture for 2.5D DNN accelerators, addressing bandwidth limitations and optimizing dataflow for enhanced performance.

Findings

WIENNA achieves 2.2X to 5.1X higher throughput.

Energy consumption is reduced by 38.2%.

Wireless NoP improves multicasting capabilities over traditional interposers.

Abstract

Deep neural network (DNN) models continue to grow in size and complexity, demanding higher computational power to enable real-time inference. To efficiently deliver such computational demands, hardware accelerators are being developed and deployed across scales. This naturally requires an efficient scale-out mechanism for increasing compute density as required by the application. 2.5D integration over interposer has emerged as a promising solution, but as we show in this work, the limited interposer bandwidth and multiple hops in the Network-on-Package (NoP) can diminish the benefits of the approach. To cope with this challenge, we propose WIENNA, a wireless NoP-based 2.5D DNN accelerator. In WIENNA, the wireless NoP connects an array of DNN accelerator chiplets to the global buffer chiplet, providing high-bandwidth multicasting capabilities. Here, we also identify the dataflow style that most efficienty exploits the wireless NoP's high-bandwidth multicasting capability on each layer. With modest area and power overheads, WIENNA achieves 2.2X--5.1X higher throughput and 38.2% lower energy than an interposer-based NoP design.