GPoS: Geospatially-aware Proof of Stake Journal: TWEB Volume: 1 1 1 1 DOI: 10.1145/3774629 CCS: Computer systems organization Peer-to-peer architectures CCS: Security and privacy Distributed systems security CCS: Networks Network reliability CCS: Networks Network performance analysis CCS: Theory of

Shashank Motepalli, Naman Garg, Gengrui Zhang, Hans-Arno Jacobsen · 2025

Read GPoS: Geospatially-aware Proof of Stake Journal: TWEB Volume: 1 1 1 1 DOI: 10.1145/3774629 CCS: Computer systems organization Peer-to-peer architectures CCS: Security and privacy Distributed systems security CCS: Networks Network reliability CCS: Networks Network performance analysis CCS: Theory of by Shashank Motepalli, Naman Garg, Gengrui Zhang, Hans-Arno Jacobsen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-02034.txt.

Introduction

Shashank Motepalli, Naman Garg, Gengrui Zhang, and Hans-Arno Jacobsen study a hidden weakness in major blockchains: their validators cluster in a few places. The team checked five large proof of stake chains: Aptos, Avalanche, Ethereum, Solana, and Sui. In each, a small set of regions held most of the voting power in consensus. That clustering, they argue, cuts against resilience, fairness, and safety from local rules.

Their fix is GPoS, short for geospatially aware proof of stake. It folds geographic diversity into stake-based voting power, so spread out validators count for more. In experiments, GPoS raised geographic decentralization by about 45 percent on average, measured with a Gini coefficient of eigenvector centrality. The cost was small: tests with the HotStuff and CometBFT protocols showed minimal extra overhead. The authors argue where validators sit should be treated as part of a chain's security.

Catalog record

CREATORShashank Motepalli, Naman Garg, Gengrui Zhang, Hans-Arno Jacobsen
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-02034

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

GPoS: Geospatially-aware Proof of Stake Journal: TWEB Volume: 1 1 1 1 DOI: 10.1145/3774629 CCS: Computer systems organization Peer-to-peer architectures CCS: Security and privacy Distributed systems security CCS: Networks Network reliability CCS: Networks Network performance analysis CCS: Theory of computation Distributed algorithms Shashank Motepalli, Naman Garg, Gengrui Zhang, Hans-Arno Jacobsen Abstract . Geospatial decentralization is essential for blockchains, ensuring regulatory resilience, robustness, and fairness. We empirically analyze five major Proof of Stake (PoS) blockchains—Aptos, Avalanche, Ethereum, Solana, and Sui—revealing that a few geographic regions dominate consensus voting power, resulting in limited geospatial decentralization. To address this, we propose Geospatially-aware Proof of Stake (GPoS), which integrates geospatial diversity with stake-based voting power. Experimental evaluation demonstrates an average 45% improvement in geospatial decentralization, as measured by the Gini coefficient of Eigenvector centrality, while incurring minimal performance overhead in BFT protocols, including HotStuff and CometBFT. These results demonstrate that GPoS can improve geospatial decentralization while, in our experiments, incurring minimal overhead to consensus performance. † † cc-license: by 1. Introduction While decentralization is a core premise for effectively and robustly operating blockchain systems ( Nakamoto 2008 ) , one critical dimension contributing to decentralization—the geospatial decentralization — remains overlooked. Geospatial decentralization refers to the geospatial distribution of validators participating in the blockchain consensus mechanisms ( Motepalli and Jacobsen 2023 ) . Geospatial centralization, i.e., clustering validators in certain regions, not only undermines decentralization but also increases vulnerability to localized risks, such as regulatory interventions, natural disasters, or targeted attacks. Key reasons for prioritizing geospatial decentralization include: (1) Regulatory and Political Control. Geospatial centralization makes blockchains vulnerable to regulatory control, where governments or authorities in specific regions may exert control over the blockchain. For example, the U.S. SEC has asserted regulatory jurisdiction over Ethereum transactions based on validator locations ( Lutz 2022 ) . Furthermore, US Treasury sanctions on certain blockchain addresses raise concerns about censorship ( U.S. Department of the Treasury 2024 ; Wahrstätter [n. d.] ) and control by authoritarian regimes, threatening the neutrality and governance of a blockchain. (2) Robustness Against Attacks and Failures: A geographically centralized blockchain network is vulnerable to region-specific failures such as natural disasters, cyberattacks, or geopolitical unrest. In 2021, an outage in an AWS data center impacted the liveness of the Solana blockchain ( Cooling 2021 ) . Additionally, centralized cloud providers can disable validators through policy changes ( Lutz [n. d.] ) , potentially halting consensus and concentrating power in centralized entities. (3) Equitable Participation and Fairness: Geospatial centralization provides latency advantages to validators in certain regions, enabling them to profit from front-running and maximal extractable value (MEV) opportunities ( Bahrani et al. 2024 ; Daian et al. 2020 ; Heimbach et al. 2023 ) . This proximity-based advantage also promotes high-frequency trading and arbitrage ( Gupta et al. 2023 ; Lewis 2014 ) , concentrating control in certain regions, skewing incentives, and increasing disparities in access to consensus. The problem this paper addresses is to drive a system towards supporting geospatial decentralization in blockchain consensus mechanisms. Widely adopted PoS (Proof of Stake) systems determine voting power in consensus based solely on staked assets, overlooking the geospatial distribution of validators. Moreover, existing decentralization metrics, such as the Nakamoto coefficient ( Motepalli and Jacobsen 2024 ; Srinivasan and Lee [n. d.] ) , fail to account for this dimension, resulting in geospatial centralization. This compromises network resilience, enables MEV exploitation ( Daian et al. 2020 ) , and threatens both blockchain neutrality and equitable global participation. Addressing this problem is challenging due to the inherent trade-offs between improving geospatial decentralization and maintaining system performance, measured in throughput and latency. To address this challenge, we design geospatially-aware consensus mechanisms. We begin by defining our system model (Section 2 ). Then, we collect empirical data on validator stake and geospatial coordinates from leading blockchains, such as Aptos, Avalanche, Ethereum, Solana, and Sui (Section 3 ). We conduct empirical analysis to quantify geospatial decentralization using the Gini coefficient of the eigenvector centrality measure. Our findings indicate significant geospatial centralization, underscoring the need for consensus mechanisms founded on more robust decentralization principles. To enhance geospatial decentralization, we propose GPoS, a mechanism that incorporates both staked assets and geospatial distribution into voting power for consensus (Section 5 ). Using our collected data, we evaluate GPoS against traditional mechanisms, demonstrating average improvements of 45% in geospatial decentralization (Section 5.6 ). We emulate validator distributions across various consensus mechanisms, including CometBFT (formerly known as Tendermint ( Buchman 2016 ) ) and HotStuff ( Yin et al. 2019 ) , to show that GPoS incurs minimal performance overhead, measured by throughput and latency (Section 6 ). The contributions of this paper are four-fold: (1) We propose GPoS , a geospatially-aware extension to stake-based voting power, enhancing decentralization in blockchain consensus. (2) We collect and analyze validator geospatial and stake data from five major blockchains, creating a comprehensive dataset that facilitates reproducibility and advances decentralization research. 1 1 1 The dataset, including raw validator geolocations, stakes, and scripts for pre-processing (e.g., proximity merging and stake aggregation), is available in the repo: https://github.com/GeoDecConsensus/geo-analysis. (3) We introduce a new metric to quantify geospatial decentralization using real-world data, providing insights into validator concentration and distribution. (4) We are among the first to empirically explore trade-offs between geospatial decentralization and performance, providing guidelines to optimize blockchain efficiency and robustness. We frame robustness via the standard correlated failure model: geographic concentration increases shared-fate risks (e.g., outages, policy actions). Dispersing voting power across regions and providers mitigates these vulnerabilities. 2. System Model Let \mathcal{V}=\{v_{1},v_{2},\ldots,v_{n}\} be the set of validators for blockchain C at epoch t , where epoch t spans approximately one day. Let sk_{i} represent the secret key for validator v_{i} , with its corresponding signature denoted as \text{sig}(sk_{i}) . We posit that cryptographic signature schemes are secure and robust, ensuring their resilience against known attack vectors ( Diffie and Hellman 2022 ; Rivest et al. 1978 ) . The voting power of validator v_{i} in the blockchain is denoted by \rho_{i} , where 0<\rho_{i}\leq 1 . The total voting power of the blockchain C is the sum of the voting powers of all validators, indicated by \sum_{v_{i}\in\mathcal{V}}\rho_{i}=1 . 2.1. Proof of Stake (PoS) Blockchains are susceptible to Sybil attacks ( Douceur 2002 ) , where a malicious actor undermines the system by masquerading as multiple validators to gain disproportionate voting power. To mitigate this risk, many blockchains use a Proof of Stake (PoS) mechanism. In PoS, voting power is proportional to the number of (native) tokens staked, referred to as stake . This model is secure due to the finite token supply, indicating a vested interest in blockchain security. Let S_{i}>0 represent the stake of validator v_{i} in the blockchain C , then normalized stake is represented by s_{i}>0 , expressed as: (1) s_{i}=\frac{S_{i}}{\sum_{\forall v_{k}\in\mathcal{V}}S_{k}} The voting power of validator v_{i} , in PoS, is its normalized stake, \rho_{i}=s_{i} . While the rest of the paper discusses PoS, the concepts are applicable to Delegated Proof of Stake (DPoS) systems, where stakeholders can delegate their stake to other validators. 2.2. Weighted Consensus The effectiveness of PoS blockchains relies on a robust consensus mechanism, which is essential for validators to agree

Read the complete text