Research | ProbeLab Analytics

MIDDLEWARE '25 Conference Paper · Dec 2025

PANDAS: Peer-to-peer, Adaptive Networking Allowing Data Availability Sampling within Ethereum Consensus Timebounds

Matthieu Pigaglio · Onur Ascigil · Michał Król · Felix Lange · Kaleem Peeroo · Sergi Rene · Ramin Sadre · Vladimir Stankovic · Etienne Rivière

Layer-2 protocols such as rollups can help address Ethereum's throughput limits. An efficient data availability layer is key for layer-2 support in Ethereum, but broadcast methods do not scale. A promising approach is the selective distribution of layer-2 data and its verification by data availability sampling (DAS). Integrating DAS with Ethereum consensus is, however, a challenge, as data must be shared and sampled within 4 seconds of each consensus slot. We propose PANDAS, a practical approach to integrating DAS with Ethereum without modifying Ethereum's core protocols. PANDAS disseminates layer-2 data and samples its availability using lightweight, direct exchanges. Its design accounts for message loss, node failures, and unresponsive participants. Our evaluation in a 1,000-node cluster and simulations for up to 20,000 peers show that PANDAS allows layer-2 data dissemination and sampling under planetary-scale latencies within the 4-second deadline.

SIGMETRICS '26 Conference Paper · Dec 2025

Multiple Sides of 36 Coins: Measuring Peer-to-Peer Infrastructure Across Cryptocurrencies

Lucianna Kiffer · Lioba Heimbach · Dennis Trautwein · Yann Vonlanthen · Oliver Gasser

Blockchain technologies underpin an expanding ecosystem of decentralized applications, financial systems, and infrastructure. However, the fundamental networking layer that sustains these systems, the peer-to-peer (P2P) layer, of all but the top few ecosystems remains largely opaque. In this paper, we present the first longitudinal, cross-network measurement study of 36 public blockchain networks. Over 9 months (since late 2024), we deployed 15 active crawlers, sourced data from two additional community crawlers, and conducted hourly connectivity probes (e.g., pings and protocol-level handshakes) to observe the evolving state of these networks. Furthermore, by leveraging Ethereum's discovery protocols, we inferred metadata for an additional 19 auxiliary networks that utilize the Ethereum peer discovery protocol. We also explored Internet-wide scans, which only require probing each protocol's default ports with a simple, network-specific payload. This approach allows us to rapidly identify responsive peers across the entire address space without having to implement custom discovery and handshake logic for every blockchain. We validated this method on Bitcoin and similar networks with known ground truth, then applied it to Cardano, which we could not crawl directly. Our study uncovers dramatic variation in network size from under 10 to more than 10,000 active nodes. We quantify trends in IPv4 versus IPv6 usage, analyze autonomous systems and geographic concentration, and characterize churn, diurnal behavior, and the coverage and redundancy of discovery protocols. These findings expose critical differences in network resilience, decentralization, and observability. Beyond characterizing each network, our methodology demonstrates a general framework for measuring decentralized networks at scale. This opens the door for continued monitoring, benchmarking, and more transparent assessments of blockchain infrastructure across diverse ecosystems.

EthResearch Technical Report · Jun 2025

Impact of IDONTWANT in the Number of Duplicates

Mikel Cortes · Yiannis Psaras

This Ethereum Research study analyzes the impact of the IDONTWANT message primitive on reducing duplicate message propagation in Ethereum's Gossipsub network following its adoption during the Pectra upgrade (with ~95% network adoption by May 2025). The research found that while IDONTWANT does provide modest improvements for larger messages like beacon blocks and blobs—increasing messages with zero duplicates from 2% to 9% and reducing average duplicates from 3 to 2—its effectiveness is significantly limited because approximately 70% of observed duplicates still arrive after IDONTWANT messages are sent, indicating that messages are already in transit by the time the control notification reaches peers. The study identified additional inefficiencies, including duplicates resulting from IWANT requests (~29% of cases) where nodes request messages that have already begun downloading, and instances where some protocol implementations fail to respect IDONTWANT signals and continue transmitting messages anyway. The researchers recommend improvements such as limiting IWANT message frequency, delaying initial IWANT requests to avoid requesting messages that are already arriving, ensuring implementations cancel IWANT replies upon receiving IDONTWANT, and fixing cases where published messages queued for transmission aren't cancelled when IDONTWANT is received.

Springer Cluster Computing Journal Article · Jun 2025

The Impact of Connectivity and Software in Ethereum Validator Performance

Mikel Cortes-Goicoechea · Tarun Mohandas-Daryanani · Jose Luis Muñoz-Tapia · Leonardo Bautista-Gomez

Modern public blockchains like Ethereum rely on p2p networks to run distributed and censorship-resistant applications. With its wide adoption, it operates as a highly critical public ledger. On its transition to become more scalable and sustainable, shifting to PoS without sacrificing the security and resilience of PoW, Ethereum offers a range of consensus client implementations to participate in the network. In this paper, we present a methodology to measure the performance of the consensus clients based on the latency to receive messages from the p2p network. The paper includes a study that identifies the incentives and limitations that the network experiences, presenting insights about the latency impact derived from running the different consensus implementations at different locations. Our study highlights the need for a holistic approach to node deployment, where hardware, software, and geographic factors have to be carefully considered. Properly dimensioned hardware is essential to mitigate latency-related performance issues and ensure the reliable operation of beacon nodes, especially in geographically distant locations.

EthResearch Technical Report · May 2025

Empirical Blob Sidecar Hit Rate Based on Local EL’s Mempool

Mikel Cortes · Yiannis Psaras

This empirical study measures blob sidecar availability in Ethereum's execution layer (EL) mempool to assess the feasibility of distributed block building for reducing validator bandwidth requirements. Analyzing data from both pre- and post-Pectra periods (which increased blob targets from 3 to 6 and maximum from 6 to 9), the researchers found a high empirical blob hit-rate where 76.6% of engine API requests successfully retrieved all requested blob sidecars from the local EL mempool within 100 milliseconds, with 98% of partial responses missing only a single blob sidecar. The study reveals that most blob sidecars are already present in the EL by the time blocks are broadcast on the consensus layer, indicating that the network is currently generating significant redundant traffic in blob propagation. While this redundancy provides resilience, it also represents a bottleneck and inefficiency; the authors recommend exploring alternatives such as shifting blob sharding to the EL layer and implementing a "Blob mempool DHT" proposal to enable more efficient network resource utilization and reduce the time pressure on consensus layer blob broadcasting.

EthResearch Technical Report · Apr 2025

Theoretical Blob Hit Rate Based on the EL Mempool

Mikel Cortes · Yiannis Psaras

This theoretical analysis by Mikel Cortes and Yiannis Psaras from ProbeLab examines blob transaction hit rates in Ethereum's execution layer (EL) mempool to assess the feasibility of distributed block-building as a mechanism to reduce bandwidth burdens on validators who build blocks locally. Using data from the public Xatu database spanning March 1-15, 2025, the study demonstrates that blob transactions achieve high visibility across the network with 99% propagating within one second across geographically dispersed regions. Key findings show that 81.91% of proposed blob transactions are available in the public mempool before the slot in which they are included, with only 4.12% arriving after slot start, and 75.11% of blocks have all their blob transactions present in the EL mempool at the time of proposal. The results indicate that distributed block-building is theoretically viable at current blob transaction rates, though the analysis also reveals that the network currently propagates redundant information by broadcasting blob transactions first through the EL mempool and then rebroadcasting sidecars over the consensus layer's GossipSub topics, while approximately 14.76% of proposed blob transactions originate from private mempools.

IEEE Access V12 Journal Article · Mar 2025

Can We Run Our Ethereum Nodes at Home?

Mikel Cortes-Goicoechea · Tarun Mohandas-Daryanani · Jose Luis Muñoz-Tapia · Leonardo Bautista-Gomez

Scalability is a common issue among the most used permissionless blockchains, and several approaches have been proposed to solve it. However, tackling scalability while preserving the security and decentralization of the network is a significant challenge. To deliver effective scaling solutions, Ethereum achieved a significant protocol improvement, including a change in the consensus mechanism towards Proof of Stake. This improvement greatly reduced the hardware requirements to run a node, leading to significant sustainability benefits with a lower network energy consumption. This work analyzes the resource usage behaviour of different clients running as Ethereum consensus nodes, comparing their performance under different configurations and analyzing their differences. Our results show higher requirements than initially claimed and show how different clients react to network perturbations. Furthermore, we discuss the differences between the consensus clients, including their strong points and limitations.