TOP 15 Protocols and Consensus Algorithms in Blockchain

What is a consensus algorithm?

Consensus is very important in all areas of our life, using it, the parties involved are ready to make decisions on important issues. Blockchain systems are no exception as they use consensus mechanisms to allow participants to reach an agreement on the network. Without a consensus algorithm, the blockchain network is nothing more than an immutable database.

Blockchain projects are decentralized and have no central authority on the network, so consensus algorithms are important to them. It allows nodes to collectively bargain with each other, motivates participants to abide by rules, ensures fairness, avoids issues such as double spending, and ensures network resiliency. The consensus mechanism remains secure and functional

How do consensus algorithms work?

Consensus algorithms are vital in large-scale fault-tolerant systems because they allow a set of distributed/replicated machines or servers to work as a single group and agree on the state of the system even in the event of failures or downtime. To do this, the algorithm sets a threshold or the number of participating machines that must reach a consensus or agreement. When they solve the consensus problem, the consensus algorithms assume that some processes and systems will not be available and that only a fraction of the nodes will respond. They also assume that some messages will be lost during transmission. However, a response is required from available nodes. For example, the algorithm may require that at least 51% of the nodes respond in order to reach a consensus or agreement on the meaning of the data or the state of the network.

This ensures that consensus is reached with minimal resources, even if other resources are unavailable or even out of order. The mechanism also maintains the integrity of the decisions made by the matching nodes in a fault-tolerant system.

Application of Consensus Algorithms in Blockchain

Consensus algorithms have many real-world applications in decentralized or distributed computer networks. One of the most common applications is blockchain.

Blockchain is the distributed ledger most associated with the bitcoin cryptocurrency. This decentralized database is collectively managed by distributed computers or nodes in a distributed peer-to-peer network. Each peer or node maintains a copy of the registry to prevent a single point of failure. Any online updates or checks are reflected in all copies at the same time. This guarantees the accuracy and security of data records and creates trust in the system without the need for a centralized trusted third party.

Blockchain networks rely on consensus algorithms to reach agreement among various distributed nodes. A consensus mechanism such as proof of work (PoW) or proof of stake (PoS) secures the network and prevents unauthorized users from verifying invalid transactions. This mechanism also allows the network to be negotiated even if no node is responsible.

Other Applications of Consensus Algorithms

Based on the underlying mechanism, consensus algorithms decide whether a distributed transaction should be committed to the database. They are also commonly used to synchronize data on a decentralized network and ensure the consistency and transparency of transactions. Consensus algorithms are also used to assign leader status to a node.

Consensus algorithms synchronize state machine replicas and ensure consistency between them. They are often used to provide trust and security in a decentralized computer network such as a blockchain and are very useful for record keeping.

In addition to blockchain and cryptocurrencies, these algorithms support many real computing and digital systems, including:

• load balancing,
• clock synchronization;
• control of unmanned aerial vehicles such as drones.

Overview of the main consensus algorithms

1. Proof-of-Work (PoW - Proof of Work)

Proof of Work (PoW) is the oldest and most widely used of blockchain consensus algorithms. It was first introduced in Bitcoin and combines the features of P2P networks, Merkle chains and cryptographic signatures. In Proof-of-Work, miners hash data and constantly use their computing resources to solve complex mathematical puzzles in order to validate transactions and add them to the blockchain. Miners in PoW networks compete with each other to earn block rewards and also receive a share of the transaction fee. In PoW, the protocol lays out the conditions that make a block valid. All miners are trying to find a valid hash, the only way to do this is by brute force input. For large blockchains, this hash is extremely complex. To compete with other miners, you need special hashing hardware (ASIC) to be able to generate a valid block. The network needs to make sure that the miner created a valid block. The algorithm needs to run the miner's data through a function, and if it produces a valid hash, it will be approved and the miner will be rewarded. Otherwise, the network would reject it, and the miner wasted time and energy.

These puzzles ensure that miners are allocated a significant amount of resources - a combination of time and processing power. When attackers want to damage the ecosystem, they have to spend a lot of money. For a 51% attack, attackers need very expensive hardware to take over most of the network. However, PoW cannot operate at maximum speeds and maintain an efficient and scalable network.

Usage example: Bitcoin, Litecoin.

2. Proof-of-Stake (PoS - Proof of Stake)

Due to the energy intensive nature of Proof of Work, Ethereum is currently in the process of transitioning from Proof of Work to Proof of Stake (PoS) consensus. In Proof of Stake, miners are replaced by validators. Validators place (store) a certain number of tokens of the corresponding cryptocurrency on the network. People who have staked a certain number of tokens vote to validate the block. If the majority of validators agree that the block is real, it is added to the blockchain. Thus, block verification in Proof of Stake consensus is done by voting, in contrast to Proof of Work where block verification is done by solving puzzles.

In Proof of Stake, the rule usually applies: the more tokens you stake, the more likely you are to be selected as a validator. When a validator is selected, it must propose (or "forge") a block. If this block is recognized as valid by other users, the validator receives a reward consisting of transaction fees for this block. Forging a block requires much less energy, as it requires many times less computing power from computers. This makes Proof of Stake much more sustainable than its counterpart.

Proof of Stake, unlike Proof of Work, is less secure, because with Proof of Stake, security is determined by people, not computers (who solve mathematical problems). Despite this, the Proof of Stake consensus can be considered as a secure consensus protocol. This is because the people who are validators have invested a significant amount of money in the cryptocurrency in question, so they are unlikely to sabotage the system. While the likelihood is minimal, there is a chance that on PoS a group of validators will take over to sabotage the system. To do this, more than 51% of the validators must agree to the sabotage plan. With a larger share of the vote, validators can influence the vote, also known as a “51% attack.” These attacks are extremely rare and in fact are never implemented in practice, because in the case of well-known and “more reliable” cryptocurrencies, more than 51% of tokens rarely end up in the hands of attackers.

Use case: NXT, Tezos, Ethereum.

3. Delegated Proof-of-Stake (DPoS - Delegated Proof of Stake)

The Delegated Proof-of-Stake (DPoS) consensus algorithm works in the same way as the PoS model, but the stakers in the DPoS model pass block validation to multiple delegates. These delegates secure the blockchain system by being responsible for maintaining consensus on the network while validating new blocks. These elected delegates then make important decisions for the entire network, such as determining what transactions are allowed and setting protocol rules.

Considering that the Delegated Proof of Stake mechanism is similar to Proof of Stake in the sense that both require users to bet coins as a means of participating in the consensus. The main difference is that in the PoS network, all coin holders participate in checking and making decisions directly through betting. In contrast, only the elected delegates participate in the verification in DPoS networks, while the remaining nodes participate indirectly by voting for the delegate(s) of their choice. The right to vote depends on how many coins the user bets. Users who bet more coins have a greater influence on the choice of nodes.

Each DPoS based network has its own voting system. Generally speaking, each delegate candidate submits a proposal when votes are requested. Candidates often promise, if elected, to share the block reward (coins received for confirming transactions) in proportion to the users who voted for them.

The number of elected delegates varies greatly for each blockchain network. Regardless of this number, all DPoS based networks assume that the majority of elected delegates are honest. In other words, delegates want to keep the network secure and will not deliberately try to confirm fraudulent transactions. As long as this is true, honest nodes will receive block rewards, while dishonest nodes will not.

Use case: EOS, BitShares, Tron, Cardano, Cosmos.

4. Proof-of-Authority (PoA)

Proof-of-Activity (PoA) combines the PoW and PoS protocols, which means that participants can both mine and pledge a stake to validate blocks. So, the PoA protocol provides a balance between miners and ordinary network participants.

PoA takes advantage of real identities to provide verification on the blockchain. This means that validators put their real identity and reputation as a guarantee of transparency. In addition, PoA is based on a limited number of validators. This feature gives it a distinct advantage, the high scalability of the blockchain, which is beneficial in applications where speed is paramount. In addition, it maintains a high level of access control to the specified blockchain, as only nodes with permission can participate.

The operation of the PoA protocol is quite simple. For the system to work, validators must be chosen at random. Enabling and selecting nodes is done through a voting system of other previously authorized nodes. In this way, you can avoid turning on malicious nodes and affecting the network. In addition to this, each validator can sign at most one of a series of consecutive blocks during their validation. Also, PoA doesn't require mining like Bitcoin does, so it's very eco-friendly.

Usage example: Decred.

5. Proof-of-Location (PoL - Proof of Location)

Proof-of-Location (PoL) or proof of location is a technology that allows the physical location of a device to be transmitted to the blockchain in such a way that other devices can rely on the location data without trusting the transmitting device.

Position dependent applications include supply chain, parametric insurance, KYC/AML verification, or even consumer transactions.

Current geolocation tools are great for locating a device, but fail to offer reliable location verification services. They are unidirectional and not encrypted, which makes it problematic to use GPS for smart contracts that need to be executed while respecting spatial parameters. Dynamic location confirmation enhances the autonomous network of permissionless beacons that can offer secure location verification services through time synchronization in a decentralized manner.

Usage example: FOAM, Platinum.

6. Proof-of-Importance (PoI - Proof of Importance)

PoI is based on the idea that rewards should be based on network activity, not just the number of coins. The probability of block placement depends on several factors, including balance, reputation (determined by a separate specially designed system) and the number of transactions made from and to this address.

Proof-of-Importance is another variation of the PoS algorithm developed for the NEM blockchain in 2015.

Proof of importance determines which network nodes are eligible to add a block to the block chain based on certain properties. This process is known as "harvesting". In exchange for collecting a block, nodes receive a transaction fee associated with hosting that block.

The higher the importance score of a node, the more likely it is to be chosen to collect a block and receive that reward. If you are interested in this, you should know that the NEM protocol requires an account to hold at least 10,000 XEM.

Proof of Importance seeks to take into account the overall support of the network, and as such it can be seen as a new approach to betting and game theory.

Usage example: NEM.

7. Proof-of-Elapsed-Time (PoET - Proof of Elapsed Time)

Among blockchain consensus and mining algorithms, one of the most striking algorithms is Proof of Elapsed Time (PoET), which in Spanish means Test of Elapsed Time. It is a consensus algorithm designed from the ground up for high scalability and focused on private blockchains.

The potential of PoET is focused on application in the development of blockchains as part of systems with high information traffic. For example, companies that need comprehensive audit systems to ensure that specified information remains unchanged. This applies in particular to high-tech and automated assembly lines, chemical and medical laboratories.

Proof of Elapsed Time (PoET) was developed by Intel as a time lottery-based consensus algorithm protected by cryptography. A revolutionary concept that uses every technology that Intel can provide in its processors. In fact, this technology is heavily tied to Intel products to the point that it cannot be used with warranty outside of its products.

The work of the PoET algorithm comes down to the fact that it creates a ring of trust in which a group of participants is coordinated by a controller. This controller has the task of accepting the work of people in the trust ring and verifying its correctness. To do this, the controller combines a stopwatch and a series of cryptographic tests that randomly allow participants to create blocks on the blockchain.

Once a participant is selected, it generates a block, develops a cryptographic test, and sends it to the controller. When the controller receives a test, it checks it and, if it's valid, accepts the block. Otherwise, it discards it. Given this last decision, the timer starts again so that the selection process starts again and the test is created by another participant. The process follows this cycle of repetitions, which keeps the network running indefinitely.

Usage example: Intel.

8. Leased Proof-of-Stake (LPoS - Leased Proof of Stake)

Leased Proof-of-Stake is another variation of the PoS algorithm that was developed exclusively for the Waves blockchain.

LPoS has more extensive functionality than PoS. Here, not just nodes containing a certain amount of crypto and having the right to mine the next block of the blockchain, but also the right to rent out their own balance to full nodes.

LPoS allows users to rent out their WAVES wallets and accept rewards in return. The larger the amount we transfer, the higher the chances that this full node will be chosen to create a new block. If this full node is selected for a new block, then the tenant will be rewarded with a percentage of the transaction fees charged by the full block. .

In an LPoS environment, holders themselves can choose to either run a full node or rent it out and receive a reward.

It is also important to say that the coins from the user's wallet do not disappear when they are rented out, but at the same time you will not be able to use them, for example, transfer them to someone or exchange them. But the lease itself can be canceled at any time.

Usage example: Waves, NXT

9. Proof-of-Space (PoSpace - Proof of Space)

PoSpace has another name that is also common - Proof-of-Capacity (proof of work, PoC).

PoSpace is a consensus algorithm that uses hard disk space instead of computing power to prove a block.

Miners reserve hard drive space to confirm blocks. At the start, sets of solutions to mathematical tasks are created on the hard drive, then mining takes place. The miner is no longer looking for a solution to the problem, since it is already in the memory of one of the hard drives. The larger the disk memory, the more completed solutions there can be, and this increases the chances of finding this solution by this miner.

This concept is very interesting from a technical point of view, because it promises to show a fairly high speed, but it could not get the attention of a large number of cryptocurrency projects, as others did.

Use Case: Burstcoin, SpaceMint (SpaceCoin), Chia

10. Proof-of-Burn (PoB - Proof of Burn)

Proof-of-Burn is a consensus algorithm alternative to classic PoW and PoS, using coin burning to increase value and reduce energy consumption.

At its core, burning coins means sending them to a special address with the inability to use these coins, so they are burned (destroyed). The miner creates a new block and receives a reward in the form of new network coins. The more coins that are burned, the better the chances of getting a reward.

The purpose of Proof-of-Burn (PoB) is "negative mining". That is, to create a new token, the existing tokens are first destroyed. And this makes the process of mining a new token more difficult, because it requires investments to burn coins. But there is an upside here, as new tokens do not depreciate as quickly due to inflation.

Quite often, PoB is used in projects that issue their own tokens, because this protects them from devaluation, as well as inflation.

Usage example: Slimcoin, Counterparty, Factom.

11. Delegated Byzantine Fault Tolerance (DBFT - Delegated Byzantine Generals Task Protocol)

DBFT is a DPoS blockchain consensus mechanism focused on BFT.

As in DPoS systems, stakeholders use delegates called accountants to reach consensus and create new blocks. In addition, owners vote for their accountant in real time rather than after a fixed period. These accountants must use their real identities to provide stakeholder votes. There is no anonymity in DBFT blockchains, the level of centralization remains high, but it provides fast block generation, high scalability and energy efficiency.

Usage example: NEO, TON.

12. Practical Byzantine Fault Tolerance (PBFT - Byzantine Generals Task Protocol Implementation)

PBFT is ideal for those who are looking for a fast, scalable, yet private blockchain. The Practical Byzantine Fault Tolerance protocol is quite similar to Delegated Byzantine Fault Tolerance, namely in its centralized nature. The big difference between these two protocols is that the first one is easier to implement, and quite often well-known participants work with it in the private sector.

The validator, when receiving a message, is forced to decide whether it is true or not. The process of this verification begins with personal checks within the validator, and after that it polls all the remaining nodes in turn, asking if they believe in the authenticity of the transaction. In the case when two-thirds of the participants approve the transaction, the node can accept it and transfer its decision to the network for other validators. In such a case, consensus will be reached based on the confirmation submitted by all validators.

Practical Byzantine Fault Tolerance is best suited for systems with low latency, but at the same time it will be very sensitive to the number of validators and throughput, because 1 notification creates a large number of different checks and requests. PBFT is ideal for private use, where a heavy load is not needed, but still there is a need for a large number of transactions.

Usage example: Hyperledger, Chain.

13. Federated Byzantine Agreement (FBA - Federated Byzantine Agreement)

The protocol called the Federated Byzantine Agreement does not ask for eligibility or previously led membership, unlike the Practical Byzantine Fault Tolerance protocol and other BFT options. FBA gives everyone the opportunity to join the network. In the transaction protocol, transactions are validated with a set number of participants, who are selected from among those who are currently online.

According to FBA rules, there are gateways called Gateways and makers called Market-Makers. Both of these components of the protocol guarantee the integrity and liquidity of the network.

Gateways play the role of the most ordinary banks that own finances and provide their equivalent in virtual tokens. Makers maintain accounts with a large number of gateways and simultaneously in several currencies.

Usage example: Stellar, Ripple.

14. Directed Acyclic Graph (DAG - Directed Acyclic Graph)

The biggest problem with blockchain is its synchronous operation. None of the blockchains can be a parallel process to another. In the blockchain, you can edit the size and frequency of block execution, or the number of participants validating blocks, but all the same, the history of events will be tailored to the strictest linear sequence. As an alternative, the Directed Acyclic Graph protocol has been proposed, which works asynchronously, giving a significant advantage to simultaneous events.

In such systems, the protocol allows participants to add 1 block of transactions, confirm several previous ones. We can say that the more new transactions, the faster the old ones will be confirmed.

Usage example: IOTA, ByteBall.

15. HashGraph (HashGraph)

The creators of this protocol say that the blockchain is already an outdated system. They believe that the concept of Directed Acyclic Graph is a good alternative, but at the same time, the main distinguishing characteristic of HashGraph is the “gossip about gossip” protocol, in which a node acquires a set of transactions with a timestamp “knowing” by another node. In order for such an algorithm to work, it is necessary to know all the participants in the process. By synchronizing the nodes, each of the nodes stores all the details and history of obtaining this information by all nodes from the network. If a node sees in history that a particular message has been previously received and verified by many, it will no doubt be true.

But there are also some limitations. First, there are very few details of practical applications on a large scale, especially with blockchain projects that work. Secondly, there is a patent for the HashGraph technology and the license for its purchase is free. Thirdly, there is no strong community that would help to check the reliability of the protocol, vulnerability to hacker attacks, or any other technology compatibility issues.

The project has recently been updated and given a new name - Hedera Hashgraph.

Usage example: HashGraph.

Conclusion

Consensus protocols and algorithms are an integral part of distributed systems. First of all, with their help, you can achieve justice by avoiding system failures when one of the node participants fails. Secondly, a decentralized environment requires a solution that helps move forward and change the overall state, even in an environment where no one trusts anyone. Certain rules help to achieve "consensus".

In this article, we tried to consider the most popular and well-known protocols (algorithms) that are already used in dozens of projects. There are many other and more exotic protocols such as Cross Fault Tolerance (XFT), Paxos, Sieve, Raft, Proof-of-Stake-Time (PoST), etc., which we simply could not fit into this article, but we will definitely describe in subsequent publications.