Energy-Efficient Consensus Algorithms for Sustainable Blockchain Networks

Authors

  • Sofia Rossi Associate Professor, School of Data Science, Western Europe Data Science University, Madrid, Spain Author
  • Jonas Schmidt Research Scientist, Department of Artificial Intelligence, European Institute of AI, Berlin, Germany Author

Keywords:

energy-efficient consensus, proof-of-stake, sustainable blockchain, carbon footprint, proof-of-work, embodied energy, green blockchain, consensus energy measurement

Abstract

Bitcoin consumes roughly as much electricity as a mid-sized European country. That single fact has done more to shape public and regulatory attitudes toward blockchain than any technical argument about decentralisation or trustlessness. Ethereum's September 2022 merge to proof-of-stake cut its energy footprint by an estimated 99.95%, proving that the link between blockchain and wasteful energy consumption is a design choice, not an inevitability. But proof-of-stake is only one option in a rapidly expanding landscape of energy-efficient consensus designs. We present the Energy-Efficient Consensus Assessment Framework (EECAF), a systematic measurement and comparison of eight consensus algorithms -- proof-of-work, proof-of-stake, delegated PoS, proof-of-authority, Raft, HotStuff BFT, Avalanche, and DAG-based consensus -- evaluated on energy consumption per transaction, per-node power draw, carbon intensity, and the often-overlooked embodied energy of validator hardware. We introduce the Sustainable Consensus Index (SCI) combining energy efficiency with security guarantees, throughput, and decentralisation. Our measurements show a 127,000-fold energy gap between the least and most efficient mechanisms. DAG-based consensus achieves the highest SCI (0.924) by combining low energy consumption with strong throughput and Byzantine resilience. We also find that embodied hardware energy -- which almost no existing analysis accounts for -- adds 18 to 34% to the true energy cost of proof-of-stake and BFT networks.

Author Biographies

  • Sofia Rossi, Associate Professor, School of Data Science, Western Europe Data Science University, Madrid, Spain

    Associate Professor, School of Data Science, Western Europe Data Science University, Madrid, Spain

  • Jonas Schmidt, Research Scientist, Department of Artificial Intelligence, European Institute of AI, Berlin, Germany

    Research Scientist, Department of Artificial Intelligence, European Institute of AI, Berlin, Germany

Downloads

Published

2024-03-28

How to Cite

Energy-Efficient Consensus Algorithms for Sustainable Blockchain Networks. (2024). Blockchain, Web3 & Digital Trust Journal P-ISSN 3117-597X and E-ISSN 3117-5988, 1(1), 44-53. https://galaxiauniverse.com/index.php/BWDTJ/article/view/441