analiza aiqbench
This research addresses a critical bottleneck in edge AI by introducing an adaptive framework that dynamically partitions neural network layers across heterogeneous hardware. By leveraging real-world testbeds rather than simulations, the study demonstrates significant improvements in energy efficiency and latency over static methods for standard convolutional models like VGG16 and MobileNetV2. The work advances practical deployment strategies for resource-constrained IoT environments where runtime dynamics dictate performance requirements.
Adaptive AI Task Partitioning and Safe Offloading in Heterogeneous Edge-Cloud Continuum
arXiv:2605.09623v2 Announce Type: replace-cross Abstract: In recent years, the use of artificial intelligence on resource-constrained IoT devices has grown significantly. However, existing approaches to AI task partitioning and offloading across the edge-cloud continuum typically rely on static methods that ignore runtime dynamics. Furthermore, they are often evaluated in simulated environments rather than on real hardware. To address this gap, we propose a framework that dynamically splits neural network layers across the heterogeneous continuum. The framework profiles the model at startup, measures network link conditions between nodes, and periodically re-evaluates the partition to adapt to environmental changes. We created a physical testbed comprising a Raspberry Pi edge device, a laptop fog, and a high-performance desktop PC as the cloud. We evaluated the framework over three widely adopted convolutional neural networks: VGG16, AlexNet, and MobileNetV2. Our results show that the framework achieves reductions in energy and end-to-end latency of 27.09--35.82% and 6.34--22.92%, respectively, compared to a static partitioning baseline. These findings confirm the superiority of adaptive to static partitioning.
Abstract:In recent years, the use of artificial intelligence on resource-constrained IoT devices has grown significantly. However, existing approaches to AI task partitioning and offloading across the edge-cloud continuum typically rely on static methods that ignore runtime dynamics. Furthermore, they are often evaluated in simulated environments rather than on real hardware. To address this gap, we propose a framework that dynamically splits neural network layers across the heterogeneous continuum. The framework profiles the model at startup, measures network link conditions between nodes, and periodically re-evaluates the partition to adapt to environmental changes. We created a physical testbed comprising a Raspberry Pi edge device, a laptop fog, and a high-performance desktop PC as the cloud. We evaluated the framework over three widely adopted convolutional neural networks: VGG16, AlexNet, and MobileNetV2. Our results show that the framework achieves reductions in energy and end-to-end latency of 27.09--35.82% and 6.34--22.92%, respectively, compared to a static partitioning baseline. These findings confirm the superiority of adaptive to static partitioning.
Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI); Performance (cs.PF)
arXiv:2605.09623 [cs.DC]
(or arXiv:2605.09623v2 [cs.DC] for this version)
https://doi.org/10.1007/978-3-032-35576-8_23 DOI(s) linking to related resources
Text integral sub licența deschisă a publicației · 195 cuvinte
Subiecte conexe
News
Quantum Computing Report · 25 iul. 2026
Multi-Timescale Latent-Action DRL for Joint Optimization in Edge-Cloud Networks
arXiv · 22 iul. 2026
News Archive 2026
Quantum Computing Report · 25 iul. 2026
Building Federated Multimodal AI Workflows with NVIDIA FLARE
NVIDIA · 19 aug. 2026