SlimCaching: Edge Caching of Mixture-of-Experts for Distributed Inference
This addresses the problem of efficient distributed inference for large MoE models in edge computing, though it appears incremental as it builds on existing MoE and caching methods.
The paper tackles the storage burden of Mixture-of-Experts (MoE) models on edge devices by optimizing expert caching across edge servers to minimize inference latency, achieving significant latency reduction in simulations compared to baselines.
Mixture-of-Experts (MoE) models improve the scalability of large language models (LLMs) by activating only a small subset of relevant experts per input. However, the sheer number of expert networks in an MoE model introduces a significant storage burden for an edge device. To address this challenge, we consider a scenario where experts are dispersed within an edge network for distributed inference. Based on the popular Top-$K$ expert selection strategy, we formulate a latency minimization problem by optimizing expert caching on edge servers under storage constraints. When $K=1$, the problem reduces to a monotone submodular maximization problem with knapsack constraints, for which we design a greedy-based algorithm with a $(1 - 1/e)$-approximation guarantee. For the general case where $K\geq1$, expert co-activation within the same MoE layer introduces non-submodularity, causing greedy methods to be ineffective. To tackle this issue, we propose a successive greedy decomposition method to decompose the original problem into a series of subproblems, with each being solved by a dynamic programming approach. Furthermore, we design an accelerated algorithm based on the max-convolution technique to obtain the approximate solution with a provable guarantee in polynomial time. Simulation results on various MoE models demonstrate that our method significantly reduces inference latency compared to existing baselines.