结合深度强化学习的区块链分片系统性能优化

doi:10.3778/j.issn.1002-8331.2203-0142

摘要/Abstract

摘要： 提高区块链系统吞吐量是广泛应用区块链的关键问题之一。针对以上问题，将分片技术应用到区块链系统中，通过使区块链并行处理事务提高区块链的吞吐量。将区块链分片选择问题建立为马尔科夫决策过程（Markov decision process，MDP），并设计了基于深度强化学习（deep reinforcement learning，DRL）的区块链分片最优选择策略（branching dueling Q-network shard-based blockchain，BDQSB）。所采用的BDQSB算法克服了传统DRL算法行为空间维度高、神经网络难以训练的缺点。仿真实验结果表明，所提出的方法可以有效降低行为空间维度，提高区块链处理事务的吞吐量和可扩展性。

关键词: 区块链, 吞吐量, 分片, 深度强化学习

Abstract: Improving the throughput of the blockchain system is one of the key issues for the widespread application of blockchain. In view of the above problems, the sharding technology is applied to the blockchain system, and the throughput of the blockchain is improved by making the blockchain process transactions in parallel. The blockchain shard selection problem is established as a Markov decision process（MDP）, and a blockchain shard optimal selection strategy（BDQSB） based on deep reinforcement learning（DRL） is designed. The adopted BDQSB algorithm overcomes the shortcomings of the traditional DRL algorithm with high behavior space dimension and slow neural network training. The simulation results show that the proposed method can effectively reduce the behavior space dimension and improve the throughput and scalability of blockchain processing transactions.

Key words: blockchain, throughput, sharding, deep reinforcement learning（DRL）

温建伟, 姚冰冰, 万剑雄, 李雷孝. 结合深度强化学习的区块链分片系统性能优化[J]. 计算机工程与应用, 2022, 58(19): 116-123.

WEN Jianwei, YAO Bingbing, WAN Jianxiong, LI Leixiao. Performance Optimization of Blockchain Sharding System Combined with Deep Reinforcement Learning[J]. Computer Engineering and Applications, 2022, 58(19): 116-123.

参考文献

[1] CONTI M，GANGWAL A，TODERO M.Blockchain trilemma solver algorand has dilemma over undecidable messages[C]//Proceedings of the 14th International Conference on Availability，Reliability and Security，2019：1-8.
[2] LUU L，NARAYANANN V，ZHENG C，et al.A secure sharding protocol for open blockchains[C]//Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security，2016：17-30.
[3] KOKORIS-KOGIAS E，JOVANOVIC P，GASSER L，et al.Omniledger：A secure，scale-out，decentralized ledger via sharding[C]//Proceedings of 2018 IEEE Symposium on Security and Privacy，2018：583-598.
[4] TEAM Z.The zilliqa technical whitepaper[J].Retrieved Sept，2017，16：2019.
[5] ZAMANI M，MOVAHEDI M，RAYKOVA M.Rapidchain：Scaling blockchain via full sharding[C]//Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security，2018：931-948.
[6] FRAN?OIS-LAVET V，HENDERSON P，ISLAM R，et al.An introduction to deep reinforcement learning[J].arXiv：1811.12560，2018.
[7] YUN J，GOH Y，CHUNG J M.DQN-based optimization framework for secure sharded blockchain systems[J].IEEE Internet of Things Journal，2020，8（2）：708-722.
[8] ZHANG J T.SkyChain：A deep reinforce ment learning-empowered dynamic blockchain sharding system[C]//Proceedings of the 49th International Conference on Parallel Processing，2020：1-11.
[9] MNIH V，KAVUKCUOGLU K，SILVER D，et al.Human-level control through deep reinforcement learning[J].Nature，2015，518：529-533.
[10] TAVAKOLI A，PARDO F，KORMUSHEV P.Action branching architectures for deep reinforcement learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2018.
[11] GERVAIS A，KARAME G O，WüST K，et al.On the security and performance of proof of work blockchains[C]//Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security，2016：3-16.
[12] SUKHWANI H，MARTíNEZ J M，CHANG X，et al.Performance modeling of PBFT consensus process for permissioned blockchain network（hyper ledger fabric）[C]//Proceedings of 2017 IEEE 36th Symposium on Reliable Distributed Systems，2017：253-255.
[13] SYTA E，JOVANOVIC P，KOGIAS E K，et al.Scalable bias-resistant distributed randomness[C]//Proceedings of 2017 IEEE Symposium on Security and Privacy，2017：444-460.
[14] SYTA E，TAMAS I，VISHER D，et al.“Keeping authorities” honest or bust “with decentralized witness cosigning”[C]//Proceedings of the 2016 IEEE Symposium on Security and Privacy，2016.
[15] WHITE C C.A survey of solution techniques for the par tially observed Markov decision process[J].Annals of Operations Research，1991，32（1）：215-230.
[16] SUTTON R S，BARTO A G.Reinforcement learning：An introduction[M].[S.l.]：MIT Press，2018.
[17] YANG Z，YANG R，YU F R，et al.Sharded blockchain for col laborative computing in the internet of things：Combined of dynamic clustering and deep reinforcement learning approach[J].IEEE Internet of Things Journal，2022（1）.
[18] LIU M，YU F R，TENG Y，et al.Performance optimization for blockchain-enabled industrial internet of things（IIoT） systems：A deep reinforcement learning approach[J].IEEE Transactions on Industrial Informatics，2019，15（6）：3559-3570.