TrimCaching: Parameter-sharing Edge Caching for AI Model Downloading

TL;DR

TrimCaching introduces a parameter-sharing edge caching framework that enhances AI model download efficiency by exploiting shared model parameters, improving cache hit ratios in mobile networks.

Contribution

The paper formulates a novel parameter-sharing model placement problem and develops approximation algorithms to optimize cache hit ratios in edge networks.

Findings

Significant improvement in cache hit ratio over traditional caching methods.

Effective approximation algorithms for the parameter-sharing caching problem.

Validation through simulations demonstrating practical benefits.

Abstract

Next-generation mobile networks are expected to facilitate fast AI model downloading to end users. By caching models on edge servers, mobile networks can deliver models to end users with low latency, resulting in a paradigm of edge model caching. In this paper, we develop a novel model placement framework, called parameter-sharing model caching (TrimCaching). TrimCaching exploits the key observation that a wide range of AI models, such as convolutional neural networks or large language models, can share a significant proportion of parameter blocks containing reusable knowledge, thereby improving storage efficiency. To this end, we formulate a parameter-sharing model placement problem to maximize the cache hit ratio in multi-edge wireless networks by balancing the fundamental tradeoff between storage efficiency and service latency. We show that the formulated problem is a submodular maximization problem with submodular constraints, for which no polynomial-time approximation algorithm exists. To tackle this challenge, we study an important special case, where a small fixed number of parameter blocks are shared across models, which often holds in practice. In such a case, a polynomial-time algorithm with a $\left(1-\epsilon\right)/2$-approximation guarantee is developed. Subsequently, we address the original problem for the general case by developing a greedy algorithm. Simulation results demonstrate that the proposed TrimCaching framework significantly improves the cache hit ratio compared with state-of-the-art content caching without exploiting shared parameters in AI models.