Ethereum Behavior Analysis with NetFlow Data

Zhenzhen Li, Jiangpan Hou, Hai Wang, Chencheng Wang, Cuicui Kang, Peipei Fu

Abstract: Ethereum is a blockchain platform that can run smart contracts and implement decentralized applications over a peer-to-peer(P2P) network. As the second-largest cryptocurrency in the world, Ethereum has attracted a lot of attention from industry and academia. Researches on Ethereum and other blockchain platforms focus on application, smart contracts, and P2P networks. Most of works use the data collected by active crawling, and rarely obtained data from the passive monitoring perspective, especially in the blockchain P2P network analysis. In this work, we propose a passive method using traffic association and machine learning to conduct online Ethereum node detection in NetFlow data, and monitor the Ethereum nodes in NetFlow traffic to gather nodes connection dataset. Based on the dataset, multi-dimensional measurement and analysis is made in order to reveal the true performance of Ethereum network.