PDPU: An Open-Source Posit Dot-Product Unit for Deep Learning Applications

TL;DR

This paper introduces PDPU, an open-source, resource-efficient posit dot-product unit designed for deep learning, achieving significant reductions in area, latency, and power over existing implementations.

Contribution

The paper presents a novel, high-throughput posit dot-product hardware unit with a fused, mixed-precision architecture and a configurable generator for diverse DNN needs.

Findings

Reduces hardware area by up to 43%

Lowers latency by up to 64%

Decreases power consumption by up to 70%

Abstract

Posit has been a promising alternative to the IEEE-754 floating point format for deep learning applications due to its better trade-off between dynamic range and accuracy. However, hardware implementation of posit arithmetic requires further exploration, especially for the dot-product operations dominated in deep neural networks (DNNs). It has been implemented by either the combination of multipliers and an adder tree or cascaded fused multiply-add units, leading to poor computational efficiency and excessive hardware overhead. To address this issue, we propose an open-source posit dot-product unit, namely PDPU, that facilitates resource-efficient and high-throughput dot-product hardware implementation. PDPU not only features the fused and mixed-precision architecture that eliminates redundant latency and hardware resources, but also has a fine-grained 6-stage pipeline, improving computational efficiency. A configurable PDPU generator is further developed to meet the diverse needs of various DNNs for computational accuracy. Experimental results evaluated under the 28nm CMOS process show that PDPU reduces area, latency, and power by up to 43%, 64%, and 70%, respectively, compared to the existing implementations. Hence, PDPU has great potential as the computing core of posit-based accelerators for deep learning applications.