SatDL: Jointly Optimizing Data Redistribution and Training for Satellite-Based Distributed Learning
cs.DC, cs.LG
Submitted: 2026-08-25
Updated: 2026-08-31
Comments: Withdrawn because this version was submitted prematurely, before all co-authors had completed their review and approved the manuscript for public dissemination. As a result, this version does not represent a manuscript approved by all authors
License: http://creativecommons.org/licenses/by/4.0/
The gist: Satellite-based distributed learning promises to train machine-learning models directly in orbit using massive, globally dispersed sensor data, thereby avoiding large-scale data downloads to ground
Terminology
Abstract
Satellite-based distributed learning promises to train machine-learning models directly in orbit using massive, globally dispersed sensor data, thereby avoiding large-scale data downloads to ground servers. However, training convergence is significantly slowed by severe non-IID data, specifically label imbalance, as each satellite observes different geographic regions with distinct labels. This imbalance extends training duration and increases energy consumption for solar-powered satellites. Existing approaches either fully redistribute data to enforce IID conditions - accelerating convergence but incurring substantial communication delays - or avoid redistribution entirely by modifying local learning algorithms to mitigate the impact of label imbalance, which, however, still prolong training and increase energy use. Both extremes result in excessive total end-to-end learning time (data-transfer delay plus training time) and thus elevated onboard energy consumption. We present SatDL, a data-redistribution framework designed to minimize total end-to-end learning time. At its core, SatDL develops a Distributor-Critic framework that jointly models and optimizes data-transfer delay and training time. Evaluations through trace-driven simulations of a 1,584-satellite Starlink constellation and hardware emulations using NVIDIA Jetson and A100 GPUs across five datasets show SatDL reduces total end-to-end learning time by up to 18.6% and onboard energy consumption by 12.23-88.00%, while maintaining inference accuracy within a few percentage points of state-of-the-art baselines.
Sources
- A Thorough Assessment of the Non-IID Data Impact in Federated Learning
- The Non-IID Data Quagmire of Decentralized Machine Learning
- Federated Optimization in Heterogeneous Networks
- Hybrid-FL for Wireless Networks: Cooperative Learning Mechanism Using Non-IID Data
- On the Convergence of Local Descent Methods in Federated Learning
- The Mapillary Traffic Sign Dataset for Detection and Classification on a Global Scale
- Flower: A Friendly Federated Learning Research Framework
- Federated Learning on Non-IID Data: A Survey
- Federated Learning with Non-IID Data
- SCAFFOLD: Stochastic Controlled Averaging for Federated Learning
- Deep Residual Learning for Image Recognition
- Communication-Efficient Learning of Deep Networks from Decentralized Data
Related papers
- iScheduler: Reinforcement Learning-Driven Continual Optimization for Large-Scale Resource Investment Problems
- SAMM: Sharded Automated Market Maker
- InferScale: GPU-Native KV Injection for Personalized LLM Serving
- Vigil: Accountable Liveness against Selective Silence
- Steelhead: Interleaving Partially Synchronous and Asynchronous Commit Rules on a Shared DAG
- Pushing CPU Speech Synthesis to the Wall: Extreme Inference Tuning under Serverless Architecture and Billing