Lightspeed Data Compute for the Space Era

TL;DR

The paper addresses the bottleneck of downlink bandwidth in Earth-observation LEO megaconstellations by proposing SpaceCoMP, a Collect-Map-Reduce framework that performs data collection, local mapping, and in-orbit reduction over an optical ISL mesh. It introduces distance-aware routing, a bipartite-matching based map-task scheduler, and a center-of-AOI reduce placement strategy, validated through simulations of Walker Delta constellations with 1,000–10,000 satellites. Key results show 8–21% distance savings in routing without extra hops, 61–79% improvement in map placement over random, and 67–72% reduction in end-to-end reduce costs under realistic compression. The approach demonstrates that orbital meshes can serve not only as communication relays but as foundations for faster, in-orbit data processing, enabling real-time, bandwidth-efficient Earth observation analytics with scalable in-space compute.

Abstract

While thousands of satellites photograph Earth every day, most of that data never makes it to the ground because downlink bandwidth simply cannot keep up. Processing data in the Low Earth Orbit (LEO) zone offers promising capabilities to overcome this limitation. We propose SpaceCoMP, a MapReduce-inspired processing model for LEO satellite mesh networks. Ground stations submit queries over an area of interest; satellites collect sensor data, process it cooperatively at light-speed using inter-satellite laser links, and return only the results. Our compute model leverages space physics to accelerate computations on LEO megaconstellations. Our distance-aware routing protocol exploits orbital geometry. In addition, our bipartite match scheduling strategy places map and reduce tasks within orbital regions while minimizing aggregation costs. We have simulated constellations of 1,000-10,000 satellites showcasing 61-79% improvement in map placement efficiency over baselines, 18-28% over greedy allocation, and 67-72% reduction in aggregation cost. SpaceCoMP demonstrates that the orbital mesh is not merely useful as a communication relay, as seen today, but can provide the foundations for faster data processing above the skies.

Figures (10)

• Figure 1: SpaceCoMP Architecture
• Figure 2: Task Processor Cost Adjacency Matrix Example
• Figure 3: Routing distance vs. constellation size. Distance optimization reduces path length by 8--10% at 53° inclination, up to 21% at 87°.
• Figure 4: Hop count comparison. Distance optimization preserves hop count while reducing physical distance.
• Figure 5: Map allocation improvement. Bipartite matching achieves 61--79% improvement over random, 18--28% over eager.