A Multifaceted Look at Starlink Performance

TL;DR

The paper tackles the challenge of understanding Starlink's global performance and the internal operations that shape real-time behavior. It employs a multi-faceted approach, combining 19.2M M-Lab measurements, 1.8M RIPE Atlas measurements, and two controlled terminals to dissect bent-pipe latency and the impact of 15 s reconfigurations. Key findings include that Starlink is broadly competitive with cellular for real-time applications, yet performance varies with ground infrastructure density and is subject to global scheduling effects that cause sub-second degradations. The work advances the field by providing a global, root-cause analysis of Starlink's performance and by releasing over 300 GB of data and scripts to support further research.

Abstract

In recent years, Low-Earth Orbit (LEO) mega-constellations have emerged as a promising network technology and have ushered in a new era for democratizing Internet access. The Starlink network from SpaceX stands out as the only consumer-facing LEO network with over 2M+ customers and more than 4000 operational satellites. In this paper, we conduct the first-of-its-kind extensive multi-faceted analysis of Starlink network performance leveraging several measurement sources. First, based on 19.2M crowdsourced M-Lab speed test measurements from 34 countries since 2021, we analyze Starlink global performance relative to terrestrial cellular networks. Second, we examine Starlink's ability to support real-time web-based latency and bandwidth-critical applications by analyzing the performance of (i) Zoom video conferencing, and (ii) Luna cloud gaming, comparing it to 5G and terrestrial fiber. Third, we orchestrate targeted measurements from Starlink-enabled RIPE Atlas probes to shed light on the last-mile Starlink access and other factors affecting its performance globally. Finally, we conduct controlled experiments from Starlink dishes in two countries and analyze the impact of globally synchronized "15-second reconfiguration intervals" of the links that cause substantial latency and throughput variations. Our unique analysis provides revealing insights on global Starlink functionality and paints the most comprehensive picture of the LEO network's operation to date.

Figures (23)

• Figure 1: Orbits of three Starlink inclinations and crowdsourced Ground Station (GS) and Point-of-Presence (PoP) locations starlink-gs-pop-unofficial. Shaded regions depict Starlink's service area.
• Figure 2: Starlink follows "bent-pipe" connectivity as traffic traverses the client-side terminal, one or more satellites via inter-sat links (ISLs), nearest ground station (GS), ingressing with the terrestrial Internet via a point-of-presence (PoP).
• Figure 3: Overview of global Starlink measurements in this study. Heatmap denotes M-Lab speedtest measurement densities. Starlink RIPE Atlas probes are shown as red circles.
• Figure 4: Field-of-view experiment setup. Dishy, deployed at a high latitude location, is obstructed by a metal shielding, which restricts its connectivity to the 70$\degree$ and 97.6$\degree$ orbits.
• Figure 5: Median of minimum RTT (in ms) of devices connected via Starlink (left) and top-3 serving ISPs (right) in the same country to the nearest M-Lab server.