{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T15:18:33Z","timestamp":1783610313948,"version":"3.55.0"},"reference-count":38,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2024,8,14]],"date-time":"2024-08-14T00:00:00Z","timestamp":1723593600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["SIGOPS Oper. Syst. Rev."],"published-print":{"date-parts":[[2024,8,14]]},"abstract":"<jats:p>To acheive maximum performance, Byzantine fault-tolerant (BFT) systems must be manually tuned when hardware, network, or workload properties change. This paper presents our vision for a reinforcement learning (RL) based Byzantine fault-tolerant (BFT) system that adjusts effectively in realtime to changing fault scenarios and workloads. We identify several variables that can impact the performance of a BFT protocol, and show how these variables can serve as features in an RL engine in order to choose the context-dependent bestperforming BFT protocol in real-time. We further outline a decentralized RL approach capable of tolerating adversarial data pollution, where nodes share local metering values and reach the same learning output by consensus.<\/jats:p>","DOI":"10.1145\/3689051.3689055","type":"journal-article","created":{"date-parts":[[2024,8,14]],"date-time":"2024-08-14T22:24:10Z","timestamp":1723674250000},"page":"15-22","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Towards Truly Adaptive Byzantine Fault-Tolerant Consensus"],"prefix":"10.1145","volume":"58","author":[{"given":"Chenyuan","family":"Wu","sequence":"first","affiliation":[{"name":"University of Pennsylvania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haoyun","family":"Qin","sequence":"additional","affiliation":[{"name":"University of Pennsylvania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohammad","family":"Javad Amiri","sequence":"additional","affiliation":[{"name":"Stony Brook University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Boon","family":"Thau Loo","sequence":"additional","affiliation":[{"name":"University of Pennsylvania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dahlia","family":"Malkhi","sequence":"additional","affiliation":[{"name":"UC Santa Barbara"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ryan","family":"Marcus","sequence":"additional","affiliation":[{"name":"University of Pennsylvania"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2024,8,14]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"https:\/\/developers.diem.com\/papers\/diemconsensus- state-machine-replication-in-the-diemblockchain\/ 2021-08--17.pdf","author":"The","year":"2021","unstructured":"The diem team. https:\/\/developers.diem.com\/papers\/diemconsensus- state-machine-replication-in-the-diemblockchain\/ 2021-08--17.pdf, 2021."},{"key":"e_1_2_1_2_1","volume-title":"The International Conference on Artificial Intelligence and Statistics, AISTATS '13","author":"Agrawal Shipra","unstructured":"Shipra Agrawal and Navin Goyal. Further optimal regret bounds for thompson sampling. In The International Conference on Artificial Intelligence and Statistics, AISTATS '13."},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/TDSC.2010.70"},{"key":"e_1_2_1_4_1","volume-title":"Symposium on Networked Systems Design and Implementation (NSDI). USENIX Association","author":"Amiri Mohammad Javad","year":"2024","unstructured":"Mohammad Javad Amiri, Chenyuan Wu, Divyakant Agrawal, Amr El Abbadi, Boon Thau Loo, and Mohammad Sadoghi. The bedrock of byzantine fault tolerance: A unified platform for bft protocols analysis, implementation, and experimentation. In Symposium on Networked Systems Design and Implementation (NSDI). USENIX Association, 2024."},{"key":"e_1_2_1_5_1","unstructured":"Elli Androulaki Artem Barger Vita Bortnikov Christian Cachin Konstantinos Christidis Angelo De Caro"},{"key":"e_1_2_1_6_1","volume-title":"The next 700 bft protocols. Transactions on Computer Systems (TOCS), 32(4):12","author":"Aublin Pierre-Louis","year":"2015","unstructured":"Pierre-Louis Aublin, Rachid Guerraoui, Nikola Kne?evi \u00b4c,Vivien Qu\u00e9ma, and Marko Vukoli\u00b4c. The next 700 bft protocols. Transactions on Computer Systems (TOCS), 32(4):12, 2015."},{"key":"e_1_2_1_7_1","first-page":"328","volume-title":"Int. Conf. on Dependable Systems and Networks (DSN)","author":"Babay Amy","year":"2019","unstructured":"Amy Babay, John Schultz, Thomas Tantillo, Samuel Beckley, Eamon Jordan, Kevin Ruddell, Kevin Jordan, and Yair Amir. Deploying intrusion-tolerant scada for the power grid. In Int. Conf. on Dependable Systems and Networks (DSN), pages 328--335. IEEE, 2019."},{"key":"e_1_2_1_8_1","first-page":"904","volume-title":"Int. Parallel and Distributed Processing Symposium","author":"Bahsoun Jean-Paul","year":"2015","unstructured":"Jean-Paul Bahsoun, Rachid Guerraoui, and Ali Shoker. Making bft protocols really adaptive. In Int. Parallel and Distributed Processing Symposium, pages 904--913. IEEE, 2015."},{"key":"e_1_2_1_9_1","first-page":"35","volume-title":"Proceedings of the First Annual ACM SIGMM Conference on Multimedia Systems, MMSys '10","author":"Barker Sean Kenneth","unstructured":"Sean Kenneth Barker and Prashant Shenoy. Empirical evaluation of latency-sensitive application performance in the cloud. In Proceedings of the First Annual ACM SIGMM Conference on Multimedia Systems, MMSys '10, pages 35--46. ACM. tex.acmid= 1730842 tex.numpages= 12."},{"key":"e_1_2_1_10_1","first-page":"199","volume-title":"NIPS '14","author":"Besbes Omar","unstructured":"Omar Besbes, Yonatan Gur, and Assaf Zeevi. Stochastic multi-armed-bandit problem with non-stationary rewards. In Advances in neural information processing systems, NIPS '14, pages 199--207."},{"key":"e_1_2_1_11_1","volume-title":"Maching Learning '96","author":"Breiman Leo","unstructured":"Leo Breiman. Bagging predictors. In Machine Learning, Maching Learning '96."},{"key":"e_1_2_1_12_1","doi-asserted-by":"crossref","unstructured":"Leo Breiman. Random forests. 45(1):5--32.","DOI":"10.1023\/A:1010933404324"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.5555\/296806.296824"},{"key":"e_1_2_1_14_1","volume-title":"Practical byzantine fault tolerance and proactive recovery. Transactions on Computer Systems (TOCS), 20(4):398--461","author":"Castro Miguel","year":"2002","unstructured":"Miguel Castro and Barbara Liskov. Practical byzantine fault tolerance and proactive recovery. Transactions on Computer Systems (TOCS), 20(4):398--461, 2002."},{"key":"e_1_2_1_15_1","volume-title":"Proceedings of the 4th ACM SIGPLAN International Workshop on Machine Learning and Programming Languages,MAPL@PLDI '20","author":"Cen Lujing","unstructured":"Lujing Cen, Ryan Marcus, Hongzi Mao, Justin Gottschlich, Mohammad Alizadeh, and Tim Kraska. Learned garbage collection. In Proceedings of the 4th ACM SIGPLAN International Workshop on Machine Learning and Programming Languages,MAPL@PLDI '20. ACM."},{"key":"e_1_2_1_16_1","volume-title":"NIPS'11","author":"Chapelle Olivier","unstructured":"Olivier Chapelle and Lihong Li. An empirical evaluation of thompson sampling. In Advances in neural information processing systems, NIPS'11."},{"key":"e_1_2_1_17_1","first-page":"277","volume-title":"Symposium on Operating Systems Principles (SOSP)","author":"Clement Allen","year":"2009","unstructured":"Allen Clement, Manos Kapritsos, Sangmin Lee, Yang Wang, Lorenzo Alvisi, Mike Dahlin, and Taylor Riche. Upright cluster services. In Symposium on Operating Systems Principles (SOSP), pages 277--290. ACM, 2009."},{"key":"e_1_2_1_18_1","first-page":"153","volume-title":"Symposium on Networked Systems Design and Implementation (NSDI)","volume":"9","author":"Clement Allen","year":"2009","unstructured":"Allen Clement, Edmund L Wong, Lorenzo Alvisi, Michael Dahlin, and Mirco Marchetti. Making byzantine fault tolerant systems tolerate byzantine faults. In Symposium on Networked Systems Design and Implementation (NSDI), volume 9, pages 153--168. USENIX Association, 2009."},{"key":"e_1_2_1_19_1","first-page":"285","volume-title":"Conf. on Computer and communications security (CCS)","author":"Dobre Dan","year":"2013","unstructured":"Dan Dobre, Ghassan Karame,Wenting Li, Matthias Majuntke, Neeraj Suri, and Marko Vukoli\u00b4c. Powerstore: Proofs of writing for efficient and robust storage. In Conf. on Computer and communications security (CCS), pages 285--298. ACM, 2013."},{"key":"e_1_2_1_20_1","first-page":"1","volume-title":"Conf. on Dependable Systems and Networks Workshop (DSN-W)","author":"Garcia Miguel","year":"2013","unstructured":"Miguel Garcia, Nuno Neves, and Alysson Bessani. An intrusion-tolerant firewall design for protecting siem systems. In Conf. on Dependable Systems and Networks Workshop (DSN-W), pages 1--7. IEEE, 2013."},{"issue":"3","key":"e_1_2_1_21_1","doi-asserted-by":"crossref","first-page":"511","DOI":"10.1109\/TDSC.2016.2593442","article-title":"A layered bft protection system for critical services","volume":"15","author":"Garcia Miguel","year":"2016","unstructured":"Miguel Garcia, Nuno Neves, and Alysson Bessani. Sieveq: A layered bft protection system for critical services. IEEE Transactions on Dependable and Secure Computing, 15(3):511--525, 2016.","journal-title":"IEEE Transactions on Dependable and Secure Computing"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/DSN.2004.1311884"},{"key":"e_1_2_1_23_1","first-page":"363","volume-title":"European conf. on Computer systems (EuroSys)","author":"Guerraoui Rachid","year":"2010","unstructured":"Rachid Guerraoui, Nikola Kne?evi\u00b4c, Vivien Qu\u00e9ma, and Marko Vukoli\u00b4c. The next 700 bft protocols. In European conf. on Computer systems (EuroSys), pages 363--376. ACM, 2010."},{"key":"e_1_2_1_24_1","first-page":"568","volume-title":"Int. Conf. on Dependable Systems and Networks (DSN)","author":"Gueta Guy Golan","year":"2019","unstructured":"Guy Golan Gueta, Ittai Abraham, Shelly Grossman, Dahlia Malkhi, Benny Pinkas, Michael K Reiter, Dragos- Adrian Seredinschi, Orr Tamir, and Alin Tomescu. Sbft: a scalable decentralized trust infrastructure for blockchains. In Int. Conf. on Dependable Systems and Networks (DSN), pages 568--580. IEEE\/IFIP, 2019."},{"key":"e_1_2_1_25_1","unstructured":"Amin Kamali Verena Kantere Calisto Zuzarte and Vincent Corvinelli. Roq: Robust query optimization based on a risk-aware learned cost model."},{"key":"e_1_2_1_26_1","first-page":"295","volume-title":"European Conf. on Computer Systems (EuroSys)","author":"Kapitza R\u00fcdiger","year":"2012","unstructured":"R\u00fcdiger Kapitza, Johannes Behl, Christian Cachin, Tobias Distler, Simon Kuhnle, Seyed Vahid Mohammadi, Wolfgang Schr\u00f6der-Preikschat, and Klaus Stengel. Cheapbft: resource-efficient byzantine fault tolerance. In European Conf. on Computer Systems (EuroSys), pages 295--308. ACM, 2012."},{"issue":"1","key":"e_1_2_1_27_1","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1109\/TSG.2013.2269541","article-title":"Survivable scada via intrusion-tolerant replication","volume":"5","author":"Kirsch Jonathan","year":"2013","unstructured":"Jonathan Kirsch, Stuart Goose, Yair Amir, Dong Wei, and Paul Skare. Survivable scada via intrusion-tolerant replication. IEEE Transactions on Smart Grid, 5(1):60-- 70, 2013.","journal-title":"IEEE Transactions on Smart Grid"},{"key":"e_1_2_1_28_1","volume-title":"Zyzzyva: speculative byzantine fault tolerance. Operating Systems Review (OSR), 41(6):45--58","author":"Kotla Ramakrishna","year":"2007","unstructured":"Ramakrishna Kotla, Lorenzo Alvisi, Mike Dahlin, Allen Clement, and Edmund Wong. Zyzzyva: speculative byzantine fault tolerance. Operating Systems Review (OSR), 41(6):45--58, 2007."},{"key":"e_1_2_1_29_1","volume-title":"Tendermint: Consensus without mining","author":"Kwon Jae","year":"2014","unstructured":"Jae Kwon. Tendermint: Consensus without mining. 2014."},{"key":"e_1_2_1_30_1","volume-title":"Hotstuff-2: Optimal two-phase responsive bft. Cryptology ePrint Archive","author":"Malkhi Dahlia","year":"2023","unstructured":"Dahlia Malkhi and Kartik Nayak. Hotstuff-2: Optimal two-phase responsive bft. Cryptology ePrint Archive, 2023."},{"key":"e_1_2_1_31_1","unstructured":"Hongzi Mao Malte Schwarzkopf Shaileshh Bojja Venkatakrishnan Zili Meng and Mohammad Alizadeh. Learning scheduling algorithms for data processing clusters."},{"key":"e_1_2_1_32_1","volume-title":"Proceedings of the 2021 International Conference on Management of Data, SIGMOD '21. Award: 'best paper award'.","author":"Marcus Ryan","unstructured":"Ryan Marcus, Parimarjan Negi, Hongzi Mao, Nesime Tatbul, Mohammad Alizadeh, and Tim Kraska. Bao: Making learned query optimization practical. In Proceedings of the 2021 International Conference on Management of Data, SIGMOD '21. Award: 'best paper award'."},{"key":"e_1_2_1_33_1","unstructured":"Ian Osband and Benjamin Van Roy. Bootstrapped thompson sampling and deep exploration."},{"key":"e_1_2_1_34_1","volume-title":"Introduction to Reinforcement Learning","author":"Sutton Richard S.","unstructured":"Richard S. Sutton and Andrew G. Barto. Introduction to Reinforcement Learning. MIT Press, 1st edition.","edition":"1"},{"key":"e_1_2_1_35_1","first-page":"913","volume-title":"USENIX Security '15","author":"Varadarajan Venkatanathan","unstructured":"Venkatanathan Varadarajan, Yinqian Zhang, Thomas Ristenpart, and Michael Swift. A placement vulnerability study in {Multi-Tenant} public clouds. USENIX Security '15, pages 913--928."},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.14778\/3594512.3594531"},{"key":"e_1_2_1_37_1","unstructured":"Li Zhou. A survey on contextual multi-armed bandits."},{"key":"e_1_2_1_38_1","doi-asserted-by":"crossref","unstructured":"Rong Zhu Wei Chen Bolin Ding Xingguang Chen Andreas Pfadler Ziniu Wu and Jingren Zhou. Lero: A learning-to-rank query optimizer. 16(6):1466--1479.","DOI":"10.14778\/3583140.3583160"}],"container-title":["ACM SIGOPS Operating Systems Review"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3689051.3689055","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3689051.3689055","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T01:17:54Z","timestamp":1750295874000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3689051.3689055"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,14]]},"references-count":38,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2024,8,14]]}},"alternative-id":["10.1145\/3689051.3689055"],"URL":"https:\/\/doi.org\/10.1145\/3689051.3689055","relation":{},"ISSN":["0163-5980"],"issn-type":[{"value":"0163-5980","type":"print"}],"subject":[],"published":{"date-parts":[[2024,8,14]]},"assertion":[{"value":"2024-08-14","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}