Semantic-Aware Caching for Efficient Image Generation in Edge Computing

TL;DR

CacheGenius is a novel edge computing system that accelerates text-to-image generation by intelligently caching and retrieving semantically similar reference images, significantly reducing latency and computational costs.

Contribution

It introduces a semantic-aware caching scheme and request scheduling algorithm tailored for edge environments to improve image generation efficiency.

Findings

Reduces generation latency by 41%

Cuts computational costs by 48%

Maintains competitive image quality metrics

Abstract

Text-to-image generation employing diffusion models has attained significant popularity due to its capability to produce high-quality images that adhere to textual prompts. However, the integration of diffusion models faces critical challenges into resource-constrained mobile and edge environments because it requires multiple denoising steps from the original random noise. A practical way to speed up denoising is to initialize the process with a noised reference image that is similar to the target, since both images share similar layouts, structures, and details, allowing for fewer denoising steps. Based on this idea, we present CacheGenius, a hybrid image generation system in edge computing that accelerates generation by combining text-toimage and image-to-image workflows. It generates images from user text prompts using cached reference images. CacheGenius introduces a semantic-aware classified storage scheme and a request-scheduling algorithm that ensures semantic alignment between references and targets. To ensure sustained performance, it employs a cache maintenance policy that proactively evicts obsolete entries via correlation analysis. Evaluated in a distributed edge computing system, CacheGenius reduces generation latency by 41% and computational costs by 48% relative to baselines, while maintaining competitive evaluation metrics.