Analyzing Geospatial Distribution in Blockchains

Shashank Motepalli, Hans-Arno Jacobsen

Abstract

Blockchains are decentralized; are they genuinely? We analyze blockchain decentralization's often-overlooked but quantifiable dimension: geospatial distribution of transaction processing. Blockchains bring with them the potential for geospatially distributed transaction processing. They enable validators from geospatially distant locations to partake in consensus protocols; we refer to them as minority validators. Based on our observations, in practice, most validators are often geographically concentrated in close proximity. Furthermore, we observed that minority validators tend not to meet the performance requirements, often misidentified as crash failures. Consequently, they are subject to punishment by jailing (removal from the validator set) and/or slashing (penalty in native tokens). Our emulations, under controlled conditions, demonstrate the same results, raising serious concerns about the potential for the geospatial centralization of validators. To address this, we developed a solution that easily integrates with consensus protocols, and we demonstrated its effectiveness.

Introduction

Decentralization is the cornerstone of blockchain technology, yet it is typically measured through node counts or stake distribution. This research highlights a critical, often ignored dimension: geospatial distribution. While protocols theoretically allow global participation, a "geospatial centralization trap" exists. High-performance consensus protocols inadvertently punish geographically distant "minority validators" by misinterpreting network latency as hardware or connectivity failure.

Global validator distribution often shows heavy concentration in specific regions like the US and Europe, creating vulnerabilities.

In networks like Ethereum, approximately 70% of validators reside in the US and Germany. Such concentration leaves the network vulnerable to regional outages, localized cloud failures, or hostile jurisdictional regulations. This clustering is driven by "geospatial censorship by protocol": validators far from the network's geographic center suffer higher latency. Their votes often arrive after a quorum is reached, leading the protocol to mark them as having low "liveliness." This results in penalties like slashing or jailing, creating an evolutionary pressure that forces validators to migrate to central data centers.

Methodology

To quantify and address this bias, we developed the following tools:

  1. GeoDec Emulator: A Python-based consensus emulator that uses netem to simulate real-world network latencies based on a dataset of pairwise ping delays between 242 cities.
  2. Geospatial Diversity Index (GDI): A metric using Haversine distance to quantify a validator's distance from the rest of the set.
  3. GDIException Smart Contract: A mechanism allowing minority validators to nominate themselves as geographically distant. Using a lightweight emulator (GeoDecLite), other validators verify these claims via IP addresses to grant exceptions from jailing.

The study utilized empirical data from the 0L Blockchain (a Diem fork) and latency data from WonderProxy to calibrate the emulator.

Key Insights and Results

Our analysis confirms a "Liveliness Trap": minority validators are not failing due to hardware issues but are victims of the speed of light. In a network where two-thirds of nodes are geographically clustered, they form quorums before distant nodes can participate. We found a strong negative correlation (-0.63) between a node's GDI and its liveliness.

Empirical evidence from the 0L network showed that functional nodes in Singapore and Australia consistently failed liveliness thresholds (l < 5%) and were eliminated. Our GeoDec simulations replicated these results perfectly; when the majority of nodes were placed in the US and Finland, minority nodes in Singapore showed 0% liveliness.

The GDIException workflow allows minority validators to be verified and exempted from unfair penalties, preserving network diversity.

By implementing the GDIException contract, we demonstrated that the number of unfairly jailed minority validators could be significantly reduced. In a 64-node emulation, the system effectively identified and protected these nodes, maintaining the network's global footprint.

Conclusion

Current blockchain reward structures provide an inadvertent incentive for centralization. To maximize profits, rational validators must move hardware to the same regions as the majority. While our work focuses on preserving existing diversity by preventing unfair punishment, future research should explore active promotion of diversity by modifying validator selection processes to ensure a geographically diverse set from the outset.

Summary updated

BibTeX
@inproceedings{motepalli2023analyzing,
  author = {Shashank Motepalli and Hans-Arno Jacobsen},
  title = {{Analyzing Geospatial Distribution in Blockchains}},
  booktitle = {DAPPS},
  year = {2023},
  doi = {10.1109/dapps57946.2023.00022},
}
Figure