Search papers, labs, and topics across Lattice.
This paper introduces Iapetus, a content-aware hierarchical scheduler designed for optimizing collaborative inference of Vision Transformers (ViTs) in low Earth orbit (LEO) satellite networks. By intelligently partitioning model layers and implementing token compression based on input content and operational constraints, Iapetus significantly reduces latency and energy consumption while maintaining quality. The results demonstrate a 91.6% task completion rate at 5 tasks/s, outperforming the leading baseline, MARATD3, by 26.1 percentage points, while also achieving substantial reductions in mean latency and battery usage.
Iapetus achieves a remarkable 91.6% task completion rate while slashing mean latency and battery draw by over 50% in satellite-based ViT inference.
Collaborative inference pools distributed resources to run compute-intensive Vision Transformers (ViTs) in satellite edge computing. Model partitioning enables such collaboration by assigning consecutive layer groups to different nodes, but the large volume of intermediate activation data incurs substantial transfer overhead that can erase its benefit. Token compression reduces downstream computation and activation transfer, but its quality impact depends on input content, model depth, and earlier pruning decisions, while layer offloading must adapt to time-varying contact and battery conditions. We present \sys, a content-aware hierarchical scheduler that screens constellation-wide options to retain a bounded candidate set, then refines each candidate into a complete token compression and layer offloading trajectory using quality prediction and joint planning. A unified objective balances per-task latency, energy, and quality loss against accumulated workload and battery pressures. We implement \sys on an NVIDIA Jetson AGX Orin hardware-in-the-loop testbed and use its validated execution model for constellation-scale trace replay across multiple ViT workloads and constellation settings. At \(5\)~tasks/s, \sys accomplishes 91.6\% of released tasks, 26.1 percentage points above MARATD3, the strongest baseline, while reducing mean latency and battery draw by 53.0\% and 70.8\%, respectively, and meeting quality targets.
