Flexible Bit-Truncation Memory for Approximate Applications on the Edge

TL;DR

This paper introduces a flexible bit-truncation memory that dynamically adjusts data precision at runtime, significantly improving power efficiency for various approximate applications on edge devices.

Contribution

It presents a novel, adaptable bit-truncation memory design that supports multiple data bit-widths at runtime, unlike prior fixed or application-specific solutions.

Findings

Supports three video application types with up to 47.02% power savings

Achieves up to 51.69% power reduction in deep learning models

Low implementation cost with 2.89% silicon area overhead

Abstract

Bit truncation has demonstrated great potential to enable run-time quality-power adaptive data storage, thereby optimizing the power/energy efficiency of approximate applications and supporting their deployment in edge environments. However, existing bit-truncation memories require custom designs for a specific application. In this paper, we present a novel bit-truncation memory with full adaptation flexibility, which can truncate any number of data bits at run time to meet different quality and power trade-off requirements for various approximate applications. The developed bit-truncation memory has been applied to two representative data-intensive approximate applications: video processing and deep learning. Our experiments show that the proposed memory can support three different video applications (including luminance-aware, content-aware, and region-of-interest-aware) with enhanced power efficiency (up to 47.02% power savings) as compared to state-of-the-art. In addition, the proposed memory achieves significant (up to 51.69%) power savings for both baseline and pruned lightweight deep learning models, respectively, with a low implementation cost (2.89% silicon area overhead).