[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to main content

Advertisement

Springer Nature Link
Log in

Increasing energy efficiency of bitcoin infrastructure with reinforcement learning and one-shot path planning for the lightning network

  • S.I.: Adaptive and Learning Agents 2023
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

The lightning network (LN) is a technological solution designed to solve the bitcoin blockchain transaction speed problem by introducing off-chain transactions. Since LN is a sparse and highly distributed network with three predominant routing protocols, its native pathfinding algorithms can potentially find multi-hop payment paths similar from the payment sender’s perspective, but the algorithms themselves have different performance, computational cost, energy consumption, and ultimately different CO2 emissions per step in the pathfinding phase. Bitcoin itself generates approximately 61.4 million tons of CO2 eq. per year. Since the LN is built on top of bitcoin, every small change in its energy consumption can have a significant impact on overall pollution. In this paper, we show that the reinforcement learning (RL) approach can reduce these costs and achieve better performance in terms of energy consumption at each pathfinding step. We introduce one-shot path prediction and propose a RL solution for a network agent that learns its neighborhood and uses local knowledge to cleverly solve the pathfinding problem and outperform native pathfinding algorithms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
£29.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (United Kingdom)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data and code availability

All data and algorithms used can be found in the public repository: https://github.com/ellariel/ln-one-shot-path-planning.

Notes

  1. Each node on the path other than the sender will have to lock the payment amount on its outgoing channel in the payment for a maximal duration of lock time. Nodes can decide on the lock time of their channel. Lower lock times indicate a higher risk for the node not to be able to relay information within a time interval. Higher lock times indicate that the maximal latency is higher. So, senders prefer lower lock times, while intermediaries might prefer the increased security of longer lock times.

  2. The success rate in this case shows the relative number of transactions that were successfully routed by the agent in their neighborhood.

References

  1. Badea L, Mungiu-Pupazan MC (2021) The economic and environmental impact of bitcoin. IEEE Access 9:48091–48104. https://doi.org/10.1109/ACCESS.2021.3068636

    Article  Google Scholar 

  2. Mora C, Rollins RL, Taladay K, Kantar MB, Chock MK, Shimada M, Franklin EC (2018) Bitcoin emissions alone could push global warming above 2 °C. Nat Clim Change 8:931–933. https://doi.org/10.1038/s41558-018-0321-8

    Article  Google Scholar 

  3. Masanet E, Shehabi A, Lei N, Vranken H, Koomey J, Malmodin J (2019) Implausible projections overestimate near-term bitcoin co2 emissions. Nat Clim Chang 9:653–654. https://doi.org/10.1038/s41558-019-0535-4

    Article  Google Scholar 

  4. Stoll C, Klaaßen L, Gallersdörfer U (2019) The carbon footprint of bitcoin. Joule 3(7):1647–1661

    Article  Google Scholar 

  5. Kohli V, Chakravarty S, Chamola V, Sangwan KS, Zeadally S (2023) An analysis of energy consumption and carbon footprints of cryptocurrencies and possible solutions. Digit Commun Netw 9(1):79–89. https://doi.org/10.1016/j.dcan.2022.06.017

    Article  Google Scholar 

  6. Bedford Taylor M (2017) The evolution of bitcoin hardware. Computer 50(9):58–66. https://doi.org/10.1109/MC.2017.3571056

    Article  Google Scholar 

  7. Wang Y-Z, Wu J, Chen S-H, Chao MC-T, Yang C-H (2019) Micro-architecture optimization for low-power bitcoin mining asics. In: 2019 international symposium on vlsi design, automation and test (VLSI-DAT), pp 1–4. https://doi.org/10.1109/VLSI-DAT.2019.8741726

  8. Zamani M, Movahedi M, Raykova M (2018) Rapidchain: scaling blockchain via full sharding. In: Proceedings of the 2018 ACM sigsac conference on computer and communications security. CCS ’18, pp 931–948. Association for Computing Machinery, New York, NY, USA. https://doi.org/10.1145/3243734.3243853

  9. Zhang Y-H, Liu XF (2021) Traffic redundancy in blockchain systems: the impact of logical and physical network structures. In: 2021 IEEE international symposium on circuits and systems (ISCAS), pp 1–5. IEEE

  10. Poon J, Dryja T (2016) The bitcoin lightning network: scalable off-chain instant payments. Retrieved Feb 22, 2023 from https://lightning.network/lightning-network-paper.pdf

  11. Lightning network in-progress specifications. GitHub. Retrieved Feb 22, 2023 from https://github.com/lightning/bolts

  12. Real-time lightning network statistics. JSON data. Retrieved Feb 22, 2023 from https://1ml.com/statistics

  13. Barratt O, Scott D(2021) Comparing bitcoin & lightning energy usage to the real world. Retrieved Feb 22, 2023 from https://blog.coincorner.com/comparing-bitcoin-lightning-energy-usage-to-the-real-world-2d64c62b1783

  14. Prihodko P, Zhigulin SN, Sahno M, Ostrovskiy AB, Osuntokun O (2016) Flare : an approach to routing in lightning network white paper. Retrieved Feb 22, 2023 from https://bitfury.com/content/downloads/whitepaper_flare_an_approach_to_routing_in_lightning_network_7_7_2016.pdf

  15. Finding routes in the lightning network. Builder’s guide. Retrieved Feb 22, 2023 from https://docs.lightning.engineering/the-lightning-network/pathfinding/finding-routes-in-the-lightning-network

  16. Pacia C (2015) Lightning network skepticism. Retrieved Feb22, 2023 from https://chrispacia.wordpress.com/2015/12/23/lightning-network-skepticism

  17. Spoke-hub distribution paradigm. Wikipedia. Retrieved Feb 22, 2023 from https://en.wikipedia.org/wiki/Spoke%E2%80%93hub_distribution_paradigm

  18. Lin J-H, Marchese E, Tessone CJ, Squartini T (2022) The weighted bitcoin lightning network. Chaos, Solitons Fractals 164:112620. https://doi.org/10.1016/j.chaos.2022.112620

    Article  Google Scholar 

  19. The lightning network daemon. GitHub. Retrieved Feb 22, 2023 from https://github.com/lightningnetwork/lnd

  20. Core Lightning (CLN): a specification compliant lightning network implementation in C. GitHub. Retrieved Feb 22, 2023 from https://github.com/ElementsProject/lightning

  21. Eclair (French for lightning) is a scala implementation of the lightning network. GitHub. Retrieved Feb 22 2023 from https://github.com/ACINQ/eclair

  22. Zabka P, Foerster K-T, Schmid S, Decker C (2022) Empirical evaluation of nodes and channels of the lightning network. Pervasive Mob Comput 83:101584. https://doi.org/10.1016/j.pmcj.2022.101584

    Article  Google Scholar 

  23. Seres IA, Gulyás L, Nagy DA, Burcsi P (2019) Topological analysis of bitcoin’s lightning network. arXiv. https://doi.org/10.48550/ARXIV.1901.04972

  24. Bere, F, Seres IA, Benczur AA (2019) A cryptoeconomic traffic analysis of bitcoin’s lightning network. arXiv. https://doi.org/10.48550/ARXIV.1911.09432

  25. Shell B (2022) How many transactions can the lightning network handle? Retrieved Feb 22, 2023 from https://voltage.cloud/blog/bitcoin-education/how-many-transactions-can-the-lightning-network-handle

  26. Helseth A (2021) The state of lightning. Arcane Research Report. Retrieved Feb 22, 2023 from https://arcane.no/research/the-growth-of-the-lightning-network

  27. Sutton RS, Barto AG (2018) Reinforcement learning: an introduction, 2nd edn. The MIT Press, Cambridge

    Google Scholar 

  28. Schulman J, Wolski F, Dhariwal P, Radford A, Klimov O (2017). Proximal policy optimization algorithms arXiv. https://doi.org/10.48550/ARXIV.1707.06347

  29. Skrynnik A, Andreychuk A, Yakovlev K, Panov A (2022) Pathfinding in stochastic environments: learning vs planning. Peer J Comput Sci 8:1056. https://doi.org/10.7717/peerj-cs.1056

    Article  Google Scholar 

  30. Asgari K, Mohammadian AA, Tefagh M (2022) DyFEn: agent-based fee setting in payment channel networks. arXiv. https://doi.org/10.48550/ARXIV.2210.08197

  31. The Proximal Policy Optimization algorithm. Stable Baselines3 Library. Retrieved Feb 22, 2023 from https://stable-baselines3.readthedocs.io/en/master/modules/ppo.html (2021)

  32. Brockman G, Cheung V, Pettersson L, Schneider J, Schulman J, Tang J, Zaremba W (2016). OpenAI gym arXiv. https://doi.org/10.48550/arXiv.1606.01540

  33. Lin J-H, Primicerio K, Squartini T, Decker C, Tessone CJ (2020) Lightning network: a second path towards centralisation of the bitcoin economy. New J Phys 22(8):083022. https://doi.org/10.1088/1367-2630/aba062

    Article  Google Scholar 

  34. Rohrer, E., Malliaris, J., Tschorsch, F (2019) Discharged payment channels: quantifying the lightning network’s resilience to topology-based attacks. arXiv. https://doi.org/10.48550/ARXIV.1904.10253

  35. Tochner S, Schmid S, Zohar A (2019) hijacking routes in payment channel networks: a predictability tradeoff. arXiv. https://doi.org/10.48550/ARXIV.1909.06890

  36. Zabka P, Förster K-T, Decker C, Schmid S (2024) A centrality analysis of the lightning network. Telecommun Policy 48(2):102696. https://doi.org/10.1016/j.telpol.2023.102696

    Article  Google Scholar 

  37. Kumble SP, Roos S (2021) Comparative analysis of lightning’s routing clients. In: 2021 IEEE international conference on decentralized applications and infrastructures (DAPPS), pp 79–84. https://doi.org/10.1109/DAPPS52256.2021.00014

  38. Malavolta G, Moreno-Sanchez P, Kate A, Maffei M (2016) SilentWhispers: enforcing security and privacy in decentralized credit networks. Cryptology ePrint Archive, Paper 2016/1054. Retrieved Feb 22, 2023 from https://eprint.iacr.org/2016/1054

  39. Roos S, Moreno-Sanchez P, Kate A, Goldberg I (2017) Settling payments fast and private: efficient decentralized routing for path-based transactions. arXiv. https://doi.org/10.48550/ARXIV.1709.05748

  40. Wang, P., Xu, H., Jin, X., Wang, T.: Flash: Efficient dynamic routing for offchain networks. In: Proceedings of the 15th International Conference on Emerging Networking Experiments And Technologies. CoNEXT ’19, pp. 370–381. Association for Computing Machinery, New York, NY, USA (2019). doi: https://doi.org/10.1145/3359989.3365411

  41. Mammeri Z (2019) Reinforcement learning based routing in networks: review and classification of approaches. IEEE Access 7:55916–55950. https://doi.org/10.1109/ACCESS.2019.2913776

    Article  Google Scholar 

  42. Godfrey D, Kim B-S, Miao H, Shah B, Hayat B, Khan I, Sung T-E, Kim K-I (2021) Q-learning based routing protocol for congestion avoidance. Comput Mater Cont 68(3):3671–3692. https://doi.org/10.32604/cmc.2021.017475

    Article  Google Scholar 

  43. Davis V, Harrison B (2022) Learning a scalable algorithm for improving betweenness in the lightning network. In: 2022 fourth international conference on blockchain computing and applications (BCCA), pp 119–126. https://doi.org/10.1109/BCCA55292.2022.9922233

  44. D’Angelo G, Severini L, Velaj Y (2016) On the maximum betweenness improvement problem. Electronic notes in theoretical computer science. In: Proceedings of ICTCS 2015, the 16th Italian conference on theoretical computer science, vol 322, pp 153–168. https://doi.org/10.1016/j.entcs.2016.03.011

  45. Kulvicius T, Herzog S, Lüddecke T, Tamosiunaite M, Wörgötter F (2020). One-shot path planning for multi-agent systems using fully convolutional neural network. https://doi.org/10.48550/ARXIV.2004.00568

  46. Valko D, Kudenko D (2023) Increasing energy efficiency of bitcoin infrastructure with reinforcement learning and one-shot path planning for the lightning network. In: Proc. of the Adaptive and Learning Agents Workshop (ALA 2023) at AAMAS 2023, May 29-30. ALA 2023. Cruz, Hayes, Wang, Yates (eds.), London, UK. https://alaworkshop2023.github.io/

  47. Decker C (2020) Lightning network research: topology datasets. GitHub. Retrieved February 22, 2023 from https://github.com/lnresearch/topology. https://doi.org/10.5281/zenodo.4088530

  48. Rozemberczki B, Kiss O, Sarkar R (2020) Little ball of fur: a python library for graph sampling. In: Proceedings of the 29th ACM international conference on information and knowledge management (CIKM ’20), pp. 3133–3140. ACM

  49. Leskovec J, Faloutsos C (2006) Sampling from large graphs. In: Proceedings of the 12th ACM SIGKDD international conference on knowledge discovery and data mining. KDD ’06, pp. 631–636. Association for Computing Machinery, New York, NY, USA. https://doi.org/10.1145/1150402.1150479

  50. Kumble SP, Epema D, Roos, S (2021) How lightning’s routing diminishes its anonymity. In: Proceedings of the 16th international conference on availability, reliability and security. ARES 21. Association for Computing Machinery, New York, NY, USA. https://doi.org/10.1145/3465481.3465761

  51. Lottick K, Susai S, Friedler SA, Wilson JP (2019) Energy usage reports: environmental awareness as part of algorithmic accountability. arXiv. https://doi.org/10.48550/ARXIV.1911.08354

  52. Vries AWG (2018) Bitcoin’s growing energy problem. Joule 2:801–805

    Article  Google Scholar 

  53. Strubell E, Ganesh A, McCallum A (2019) Energy and policy considerations for deep learning in NLP. arXiv. https://doi.org/10.48550/ARXIV.1906.02243

  54. Posani L, Paccoia A, Moschettini M (2019) The carbon footprint of distributed cloud storage. arXiv. https://doi.org/10.48550/ARXIV.1803.06973

  55. Patterson D, Gonzalez J, Le Q, Liang C, Munguia L-M, Rothchild D, So D, Texier M, Dean J (2021) Carbon emissions and large neural network training. arXiv. https://doi.org/10.48550/ARXIV.2104.10350

  56. Gitzel R (2022) Software tools to determine the carbon footprint of AI code. Retrieved Feb 22, 2023 from https://www.linkedin.com/pulse/software-tools-determine-carbon-footprint-ai-code-ralf-gitzel/

  57. Schubert S, Kostic D, Zwaenepoel W, Shin KG (2012) Profiling software for energy consumption. In: 2012 IEEE international conference on green computing and communications, pp 515–522. https://doi.org/10.1109/GreenCom.2012.86

  58. Lannelongue L, Grealey J, Inouye M (2021) Green algorithms: quantifying the carbon footprint of computation. Adv Sci 8(12):2100707. https://doi.org/10.1002/advs.202100707

    Article  Google Scholar 

  59. Hessel M, Hasselt HV, Modayil J, Silver D (2019) On Inductive Biases in Deep Reinforcement Learning. arXiv. https://doi.org/10.48550/ARXIV.1907.02908

  60. Wei H, Liu X, Ying L (2023) Safe reinforcement learning with instantaneous constraints: the role of aggressive exploration. https://doi.org/10.48550/ARXIV.2312.14470

Download references

Acknowledgements

The authors would like to thank Dr.-Ing. Manfred Veenker and the Veenker Foundation for their help, and Ildar Baimuratov, Konstantin Glonin, Henrik Müller and Yuan Xue for their constructive and valuable discussions and guidance during the planning and development of this research.

Funding

Not applicable.

Author information

Authors and Affiliations

  1. OFFIS — Institute for Information Technology, Escherweg 2, 26121, Oldenburg, Lower Saxony, Germany

    Danila Valko

  2. L3S Research Center, Appelstr, 9a, 30167, Hannover, Lower Saxony, Germany

    Danila Valko & Daniel Kudenko

Authors
  1. Danila Valko
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Daniel Kudenko
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

D.V. is responsible for developing the idea, implementation, experiments, and writing article; D.K. is responsible for developing the idea, validation, supervision, and writing article.

Corresponding author

Correspondence to Danila Valko.

Ethics declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical approval

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A

Appendix A

1.1 Experiment details and hyperparameters

We used the same setup for the additional training and the experiments and employed mostly default hyperparameters for the Stable-Baselines3 PPO implementation [31]: policy type: actor-critic; timesteps per epoch: 100,000 for general training, 10,000 for experiments and additional training; learning rate: 0.0003; discount factor (\(\gamma\)): 0.99; GAE parameter (\(\lambda\)): 0.95; clipping parameter: 0.2; value function coefficient: 0.5; maximum value for the gradient clipping: 0.5.

Hardware setup for general training. CPU: AMD EPYC 7662 64-Core Processor, 256 CPUs; GPU: 2 x A100-PCIE-80GB; 1 TB RAM, 1007.764 GB available; Platform system: Linux-5.10.0-15-amd64-x86_64-with-glibc2.31; Python version: 3.10.6.

Hardware setup for experiments and additional training. CPU: 11th Gen Intel(R) Core(TM) i5-1135G7 @ 2.40GHz, 8 CPUs; GPU: 1 x NVIDIA GeForce MX350; 8 GB RAM, 7.675 GB available; Platform system: Windows-10-10.0.22621-SP0; Python version: 3.9.13.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Valko, D., Kudenko, D. Increasing energy efficiency of bitcoin infrastructure with reinforcement learning and one-shot path planning for the lightning network. Neural Comput & Applic (2024). https://doi.org/10.1007/s00521-024-10588-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00521-024-10588-2

Keywords

Search

Navigation