---
title: "Replica-Aware Split Inference for Edge IoT Systems"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.04779v1"
permalink: "https://feed.astralor.com/items/itm_b4ced11cdb3f4eb3a9f65f02"
published: "2026-11-03T12:18:16.000Z"
score: 58
channel: "tech"
tags: ["推理效率","RAG","Replica-Aware Split Inference","Edge IoT Systems","Split Inference","Internet","Things","IoT"]
---

# Replica-Aware Split Inference for Edge IoT Systems

## At a glance

Split Inference (SI) addresses the hardware resource limitations of Internet of Things (IoT) devices by distributing tiny Machine Learning (TinyML) model segments across multiple IoT devices. However, a device failure c…

## Key points

- Split Inference (SI) addresses the hardware resource limitations of Internet of Things (IoT) devices by distributing tiny Machine Learning (TinyML) model segments across multiple IoT devices.
- However, a device failure can interrupt the inference pipeline.
- This paper proposes a replica-aware recovery approach that deploys replicas of TinyML model segments originally hosted on failed devices onto surviving devices.

## Full text

Split Inference (SI) addresses the hardware resource limitations of Internet of Things (IoT) devices by distributing tiny Machine Learning (TinyML) model segments across multiple IoT devices. However, a device failure can interrupt the inference pipeline. This paper proposes a replica-aware recovery approach that deploys replicas of TinyML model segments originally hosted on failed devices onto surviving devices. We formulate the joint selection of replica hosts and communication protocols as a mixed-integer linear programming (MILP) problem that minimizes post-reconfiguration inference latency, accounting for computation and transmission latencies, under flash memory and runtime memory constraints. Simulations using MobileNetV2 on heterogeneous IoT networks compare the proposed approach with four baselines. We implement the model in a network of 15 IoT devices with a maximum link-rate reduction of 75%, and show that MILP approach achieves an average modeled end-to-end inference latency that is 6.8% lower than that of Computation Greedy and 18.3% lower than that of Random. Index Terms TinyML, split inference, Internet of Things, fault tolerance, replica placement.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_b4ced11cdb3f4eb3a9f65f02