“Draft:Delegated Proof-of-Stake”的意思、由来-开放百科全书

Delegated Proof-of-Stake (dPOS) is a consensus mechanism in distributed ledgers. DPOS is mostly associated with blockchain based systems, but not exclusively restricted to blockchain only ^[1]. Consensus mechanisms in blockchain aim to address the Byzantine fault problem. Other than proof-of-work or proof-of-stake, in delegated proof-of-stake, participation in the consensus process is permissioned and bounded to a specific set of block producers. Those block producers can be voted by the user base (like in EOS or Steem). ^[2]

History

Delegated proof-of-stake was originally invented by Daniel Larimer in April 2014 for the Graphene blockchain of Bitshares (a decentralized cryptocurrency exchange) ^[3] .

Today dPOS is used in many blockchain based systems like Bitshares 2.0, Steem, Eos, Tron, Lisk, Ark.

In contrast to proof-of-work or proof-of-stake systems like (Bitcoin or Ethereum) delegated proof-of-stake is faster ^[4], but also because of its permissioned nature, topologically more centralized than pure proof-of-Work or proof-of-Stake systems. When combined with proof-of-work, hybrids like Steem (95% dPOS and 5% POW, until hardfork 0.17 in 2017) are possible.

To date, many systems have implemented dPOS consensus mechanism. In blockchains, the delegates are sometimes called witnesses and in directed acyclic graphs stability points.How many permissioned block producers are voted is up to the protocol.

Process

N number of permissioned block producers get elected from a voted pool of block producer candidates ("Witnesses")

The block is irreversible when it is voted on by 2/3+1 of the witnesses. When not voted 2/3+1, then the "longest chain rule" is applied.

Criticism

Proof-of-work, as used in Bitcoin, has a probabilistic, not final irreversibility with synchronicity assumption and an unbounded, permissionless set of block producers (everyone can participate)

^[6]. Proof-of-work after Nakamoto 2008 has no liveness threshold, and a ½ safety-margin, which only allows 51%-attacks to be successful ^[7] (even though there are studies like Eyal and Sirer 2014 suggesting that majority alone is not enough

^[8]). Delegated proof-of-stake, on the other hand, has a bounded, synchronous finality of irreversibility, so that the bounded and permissioned set of block producers is subject to a 1/3 liveness threshold with 2/3 safety margin. This reduces (in theory) the network recourses necessary for a successful attack. Until this day, dPOS-blockchains are the most successful blockchains when it comes to use (operation per time) and peak capacity. ^[9] Some of them, like Bitshares or Steem, operate successfully for some years now ^[10].

Delegated proof-of-stake should not be confused with proof-of-stake, even though there are proof-of-stake systems with optional stake delegation like Ouroboros ^[11]

References

1. ^{{cite web|url= https://obyte.org/Byteball.pdf|title= Byteball: A Decentralized System for Storage and Transfer of Value|date= 1 October 2016 |author= A. Churyumov| accessdate=14 March 2019 }}
2. ^{{cite journal| last = Nguyen| first = Giang-Truong | last2 = Kim| first2 = Kyungbaek| date = February 2018| title = A Survey about Consensus Algorithms used in Blockchain| url = http://jips-k.org/file/down?pn=530| journal = Journal of Information Processesing Systems| volume = 14| issue = 1| pages = 101-128| doi = 10.3745/JIPS.01.0024| access-date = 14 March 2019}}
3. ^{{cite web|url= https://bitshares.org/technology/delegated-proof-of-stake-consensus/| title=Delegated Proof-of-Stake Consensus A robust and flexible decision making protocol|publisher= Bitshares|accessdate=14 March 2019}}
4. ^{{cite web|url= https://blocktivity.info/ | title= Blockchain aktivity and performace|date= 14 March 2019| publisher= blocktivity.info|accessdate=14 March 2019}}
5. ^{{cite web|url= https://medium.com/loom-network/understanding-blockchain-fundamentals-part-3-delegated-proof-of-stake-b385a6b92ef | title= Understanding Blockchain Fundamentals, Part 3: Delegated Proof of Stake|date= 11 June 2018| publisher= Georgios Konstantopoulos|accessdate=14 March 2019}}
6. ^{{cite journal| last = Miller| first = Andrew| last2 = LaViola Jr.| first2 = Joseph J.| date = April 2014| title = Anonymous Byzantine Consensus from Moderately-hard Puzzles: A Model for Bitcoin| url = https://socrates1024.s3.amazonaws.com/consensus.pdf| access-date = 14 March 2019}}
7. ^{{cite web|url= https://bitcoin.org/bitcoin.pdf | title= Bitcoin: A Peer-to-Peer Electronic Cash System|date= 2008| publisher= Satoshi Nakamoto 2008|accessdate=14 March 2019}}
8. ^{{cite journal| last = Eyal| first = Ittay | last2 = Sirer| first2 = Emin Gün| date = November 2014| title = Majority Is Not Enough: Bitcoin Mining Is Vulnerable| url = http://jips-k.org/file/down?pn=530| journal = Financial Cryptography and Data Security. FC 2014. Lecture Notes in Computer Science| volume = 8437| doi = 10.1007/978-3-662-45472-5_28| access-date = 14 March 2019}}
9. ^{{cite web|url= https://medium.com/loom-network/understanding-blockchain-fundamentals-part-3-delegated-proof-of-stake-b385a6b92ef | title= Understanding Blockchain Fundamentals, Part 3: Delegated Proof of Stake|date= 11 June 2018| publisher= Georgios Konstantopoulos|accessdate=14 March 2019}}
10. ^{{cite web|url= https://blocktivity.info/ | title= Blockchain aktivity and performace|date= 14 March 2019| publisher= blocktivity.info|accessdate=14 March 2019}}
11. ^{{cite web |url=https://eprint.iacr.org/2016/889.pdf |title=Ouroboros: A Provably Secure Proof-of-Stake Blockchain Protocol |date=21 August 2017 |author1=Aggelos Kiayias |author2=Alexander Russell |author3=Bernardo David‡ |author4=Roman Oliynykov |accessdate=14 March 2019}}