Cryptocurrencies

The DPoS the consensus that everyone seems to agree – Cryptocurrencies

Blockchain digital illuminated shape. data node base concept. 3

All blockchains are fundamentally deterministic networks updated by transactions. Consensus is the process that makes it possible to rule on a specific order of transactions and filter out invalid transactions. There are many algorithms that can produce equivalent trade orders. However, DPoS (Deleguate Proof of Stake) or delegated proof of stake has proven to be particularly robust, secure, and efficient in recent years through its use in multiple blockchains.


Community self-management

As with any other Consensus, the biggest threat a malicious third party can present is censorship. All blocks must be validated in accordance with a logical order determined by the algorithm.

The DPoS consensus is divided into two parts : choose a group of block producers and program their production over time. The process of election by vote ensures that the stakeholders (holders) of the currency are ultimately in power because they are the ones who will be at a loss if the network loses fluidity and efficiency. Indeed the validators are chosen by vote in proportion to the number of tokens that each member owns. So if I have 1 token I would have for example 10 votes, the user with 10 tokens would have 100. The way the validators are elected has little impact on how the consensus performs every minute. We are interested here in how consensus is reached on the network after the block producers have been determined. In the production of new blocks we talk about forging.

Through examples, we will determine how the DPoS works under several network conformations. These examples are used to illustrate how DPoS is sturdy and hard to break despite the forks that may arise.

Normal operation

During normal operation, block producers take turns producing one block every 3 seconds. Considering that no one misses their turn, they will thus provide the longest chain possible. If producers fail to produce a block within their allotted time interval, any other block provided outside of that interval will be invalidated.

blockchain example

Minority Fork

Until 1/3 of nodes can be malicious or malfunctioning thus creating a minority fork (bifurcation). In this case, the minority fork will only produce one block every 9 seconds while the majority fork will produce 2 blocks every 9 seconds. Once again, 2/3 of the knots, i.e. honest knots, will always have the longest chain therefore the majority channel.

fork blockchain

Double production by minority channels

The minority can try to produce an unlimited number of forks, but all these forks will be shorter than the majority chain because the minority is limited in the enlargement of the minority channel.

double fork blockchain

Network fragmentation

It is quite possible that the network is fragmented / disconnected and in this case, no fork will have a majority of producers of blocks (greater than 2 people in this example). In this case, the longest chain will then be the chain of the largest minority. When the connection to the network is restored, the smaller minorities will naturally adopt the longer chain and the ambiguity of consensus will be resolved.

example of network fragmentation

It is possible in this case that there are 3 forks with the longest chains of similar sizes. In this case, the producers of the third (smaller fork) will break the indecision when they reconnect to the network. There is an odd number of block producers therefore it is impossible to maintain indecision for a long time (there will never be a 50/50).

Later we will see that the shuffling (brewing) of block producers which makes it possible to make pseudo-random the order of passage of the producers ensures tie breaking even if two forks have the same number of block producers. This is because forks will grow at different speeds, allowing one fork to gain the upper hand over the other.

The last irreversible block

As part of a fragmentation / disconnection network, it is possible for multiple forks to continue to grow for an extended period of time. In the time, the longest chain will win, but observers demand a way to know for sure when a block is absolutely part of the fastest growing chain. This can be determined by 2/3 + 1 confirmation block producers.

In the diagram below, block B has been confirmed by C and by A which represents 2/3 + 1 of the producers, so we can deduce that no other chain would possibly be longer if at least 2/3 of the producers are honest.

Pub

Cryptoassets are highly volatile unregulated investment products. No EU investor protection. Your capital is at risk.

example of last irreversible block

This rule is similar to the 6 Confirmation Block Rule for Bitcoin (BTC). Intelligent individuals can cause a sequence of events where two nodes have two different terminal irreversible blocks. This scenario requires the attacker to havefull control of communication delay between nodes and use this control not once, but with two minutes of differences. The chances of an attack occurring on a DPoS network are close to 0 and the insignificant economic consequences.

Lack of Quorum among producers

In the rare event that the quorum (number of producers required) of producers is not satisfied, it is possible to the minority to continue to produce blocks. In these blocs, stakeholders can include transactions to change their votes. These votes can designate a new set of producers and restore participation in block production to 100%. When this happens, the minority channel may eventually gain the upper hand over all other channels operating with less than 100% ownership.

Throughout this process, all observers will be aware that the state of the network is changing before a new chain emerges with 67% participation. Those who choose to trade under these conditions, and thereby change their votes, run a similar risk to those who accept with less than 6 confirmations. They do this knowingly and know that there is still a low probability that the consensus could ultimately opt for a different fork. In practice, this situation is much more secure than accepting blocks with less than 3 Bitcoin confirmations.

Corruption of the majority of producers

If the majority of block producers become corrupted, they can produce an unlimited number of forks as well. Each of these forks will appear to progress with confirmation of the 2/3 majority. In this case, the algorithm of last unalterable block will revert to the longest chain algorithm. The longest string will be cshe approved by the broadest majority which will be decided by thea minority of the remaining honest knots.

This kind of behavior could not last because the stakeholders would quickly replace their vote towards these honest producers.

example of producer corruption

TaPoS (Transaction as Proof of Stake), transaction instead of proof of stake

When users sign a transaction, they do so with a certain guess the state of the blockchain. This hypothesis is based on their perception of recent blocks. If the longest chain consensus changes the majority chain, this could potentially invalidate the assumption made by the signer when they consented to the transaction.

With TaPoS, all transactions include a hash from a recent block and are thus considered invalid if this block does not exist in the history of the chain. Anyone who signs a transaction on an orphan fork will see their invalid transaction and unable to migrate to the main chain.

A side effect of this process is security with long range attacks that attempt to generate an alternate chain. The stakeholders involved directly confirm the blockchain each time they carry out a transaction. Over time all the blocks are confirmed by the stakeholders and this is something that cannot be reproduced in a forged chain.

Brewing of determined producers

In all the examples we have illustrated, we have shown random scheduling of block producers. In reality, the production order is shuffled every N blocks where N is the number of producers of blocks. This randomization ensures that producer B does not always ignore producer A and that at any time there are not multiple forks with the same number of producers and thus allow decision making by tie breaking.

DPoS, the perfect solution?

DPoS is robust under all forms of network disruption and even more secure in the face of corruption attempts from a large minority of producers. Contrary to competitive consensus, DPoS can continue to operate despite a majority of producers fail. During this process, the community can vote to replace defaulting producers until you find a 100% stake. Thus, DPoS is very robust against failure.

Pub

Cryptoassets are highly volatile unregulated investment products. No EU investor protection. Your capital is at risk.

Ultimately, DPoS achieves security through the algorithm it chooses to determine block producers and verify that the nodes are high quality and unique people. Using the vote-based approval process, the network ensures that even someone with 50% in the current amount of voting power is unable to choose a single producer on their own. DPoS is designed to optimize the performance of the nominal condition of 100% participation of honest nodes with a good connection. This grants DPoS the power to confirm transactions with 99.9% certainty in an average of 1.5 seconds. So many arguments that make DPoS one of the most successful consensuses in the blockchain ecosystem.

Tags

Related Articles

Back to top button
Close
Close