LEO Satellite Networks Assisted Geo-distributed Data Processing

TL;DR

This work tackles the cost-efficient transfer of geo-distributed edge-cloud data to core data centers by leveraging LEO satellite networks. It introduces DVA, a bandwidth-aware satellite selection algorithm that builds a candidate satellite bipartite graph, formulates an ILP to minimize the transmission time $T$ with binary variables $x_{i,j}$ and visibility $v_{i,j}$, and employs a greedy heuristic to achieve $O(m imes n)$ complexity while approaching the ILP optimum. Empirical results show that DVA reduces access-network transmission duration by about 50% and doubles throughput compared with baseline methods, with computation times under 1 ms and close to optimal performance. The approach is robust across multiple constellation configurations, offering a practical solution for real-time, scalable geo-distributed data processing in LEO networks, with potential extensions to distributed learning, ISAC, and LLM-enabled workloads in such networks.

Abstract

Nowadays, the increasing deployment of edge clouds globally provides users with low-latency services. However, connecting an edge cloud to a core cloud via optic cables in terrestrial networks poses significant barriers due to the prohibitively expensive building cost of optic cables. Fortunately, emerging Low Earth Orbit (LEO) satellite networks (e.g., Starlink) offer a more cost-effective solution for increasing edge clouds, and hence large volumes of data in edge clouds can be transferred to a core cloud via those networks for time-sensitive big data tasks processing, such as attack detection. However, the state-of-the-art satellite selection algorithms bring poor performance for those processing via our measurements. Therefore, we propose a novel data volume aware satellite selection algorithm, named DVA, to support such big data processing tasks. DVA first takes into account both the data size in edge clouds and satellite capacity to finalize the selection, thereby preventing congestion in the access network and reducing transmitting duration. Extensive simulations validate that DVA has a significantly lower average access network duration than the state-of-the-art satellite selection algorithms in a LEO satellite emulation platform.

Figures (5)

• Figure 1: Edge clouds access the data center via satellite network.
• Figure 2: The default selection algorithm may lead to unnecessary bandwidth competition.
• Figure 3: Edge - candidate access satellite bipartite graph.
• Figure 4: Performance Test Results of Four Algorithms
• Figure 5: The performance results among different constellations.