What Is The Federated Byzantine Agreement?

Sara Sciusco

Published in

DigitalbitsOrg

9 min readMay 16, 2022

A deeper look at the sustainability, economic incentives, and network integrity of FBA vs other consensus protocols.

Introduction

Energy usage and sustainability in the blockchain industry have been two of the hottest topics over the past year. Some of these concerns have slowed down the adoption of blockchain technology by companies. Some firms have even outright criticized blockchain technology for its energy usage. Addressing these concerns is paramount to moving closer to mass adoption.

It is key to remark that two factors that contribute the most to the energy consumption of a blockchain network are: the nodes and the consensus mechanism. These two topics work hand-in-hand. Typically they are not referenced in media pieces together. For example, some media outlets might discuss the ‘hash-rate’ of the Bitcoin blockchain. To inform the general public about the blockchain, it’s important to discuss them together. This will better address the general misconceptions around these topics.

This is why this piece was written. To explore the inter-related topics of nodes and consensus mechanisms and how they work together. This article will discuss some of the related concepts such as energy consumption, economic incentives, and network integrity.

The Consensus Mechanism Explained

At the heart of any blockchain network is the consensus mechanism. It dictates how the blockchain verifies transactions. It also contributes to the amount of energy that the blockchain network uses. Here’s an example to illustrate the point.

Imagine that you’re interested in a video game that is launching and you would only play this video game if it were considered a ‘good’ game. The only resource available to inform gamers would be reviews published by video game reviewers. The problem that arises now would be how to decide if a video game is ‘good’ based on these reviews?

One approach would be to conclude that the game is ‘good’ if there are more positive reviews than negative reviews. Alternatively, you could consider a game good if more than 90% of the published reviews were positive. There are several approaches to deciding if a video is ‘good’ based on the sentiment of reviews. The vital thing to note is that these aforementioned strategies are examples of consensus mechanisms.

Let us imagine this system in action if we decide a game is ‘good’ only if there are more positive reviews than negative reviews. Once we found an interesting video game that is launching, following this system, we would first look at all the available reviewers and observe the sentiment of their reviews. For example, if there are 100 available reviews and 90 reviews are positive, then we would conclude that the game is ‘good’. Alternatively, if 90 of the 100 reviews had a negative sentiment, we would consider the game to be ‘bad.’

This example captures the essence of how blockchains verify transactions. Instead of reviewers, we have ‘nodes.’ Instead of reviewers writing a negative or positive video game review, nodes decide if the current transactions are legitimate. And finally, the way we determine the general consensus is the consensus mechanism of the blockchain.

What is a node?

Now that we have an idea about how a consensus mechanism works, let’s discuss nodes. While the consensus mechanism decides how transactions get verified and how the nodes work together, it is the nodes that accept or reject a transaction. Consensus is reached when you aggregate a number of these individual transactions.

A node can be any computational device, from a smartphone to a supercomputer. The blockchain network is effectively a cluster of these devices, working together to verify transactions and store them in the public ledger. The amount of computational power required depends upon the complexity of the network and the consensus mechanism employed. For example, Proof-of-Work chains typically require devices with a significant amount of power. Proof-of-Stake and Federated Byzantine Agreement networks only require the node to have minimal amounts of computational power. For reference, a DigitalBits node can run on an m5.large AWS instance, which has two CPU cores and 8GB of ram.

Types of Consensus Mechanisms

Now that we understand how a consensus mechanism works, we can begin to unpack two of the most popular mechanisms and then explore the consensus mechanism that the DigitalBits blockchain network uses.

Proof-of-work (POW)

This mechanism works by having nodes compete by solving a set of mathematical equations defined by some algorithm. The first node to finish the set of equations gets to validate the transaction result and add it to the history of transactions that occurred. To incentivize nodes, to be honest, the first node that completes the equations is rewarded a set amount of tokens. If a node were to be dishonest, it would effectively diminish its reward’s value, so this reward is designed to deter dishonesty. Examples of blockchain networks that use this technology would be Ethereum, Bitcoin, and Litecoin

Proof-of-stake (POS)

This mechanism works by having token holders’ lock-up’ tokens to become a validator node. Then the network randomly chooses one of the validator nodes to verify the transaction and then commit it to the history of transactions. The probability of being selected as a validator node can be increased by staking more tokens. Like proof-of-work, chosen validators are paid a set amount of tokens upon validating a transaction, incentivizing nodes, to be honest. Additionally, the owners have to lock up their tokens, and thus dishonesty would hurt the value of these locked tokens. Examples of blockchain networks that use this would be Cardano and Avalanche.

Federated Byzantine Agreement (FBA)

The mechanism is employed by the DigitalBits Blockchain. Each node keeps a list of nodes that it trusts.’ If node A trusts node B, then node B can influence the sentiment of node A, and vice versa. Because of this ability of nodes to ‘trust’ each other, networks of nodes that trust each other build up over time. For a transaction to succeed, several nodes must agree that it is valid. To do this, a small number of nodes are chosen at random, and they check if they agree with the transaction. Then, these nodes can choose other nodes that they trust to verify the transaction. This cycle continues until the required number of nodes agree on the transaction.

Energy Consumption, Financial Incentive, and Security of Nodes

Proof-of-Work and Proof-of-Stake provide a financial incentive to operate nodes and contribute to the network’s security. However, for both Proof-of-Work and Proof-of-Stake, this results in an excess of nodes being operated, resulting in the network’s energy consumption increasing substantially.

They have to ‘mine’ for Proof-of-Work chains and thus consume immense amounts of computational energy. Furthermore, because of this reward, ‘miners’ or node operators are incentivized to deploy more computationally intensive nodes as they have to compete for rewards. This creates a cycle where node operators repeatedly deploy stronger devices to be competitive. As a consequence, the energy usage of the blockchain increases substantially throughout this cycle.

For Proof-of-Stake, the energy consumption is drastically reduced in comparison to Proof-of-Work. However, since there is still a financial incentive to run a node, nodes are unnecessarily added to the network and thus raise the network’s energy consumption. Whilst adding more nodes assists in decentralizing the network, there is a point where adding more nodes contributes very little to the integrity and stability of the network whilst unnecessarily increasing the energy usage of the network. Research done by UCL has shown that while Proof-of-Stake uses substantially less energy than Proof-of-Work, Proof-of-Stake still can use more energy than traditional payment systems.

Regarding the Federated Byzantine Agreement, nodes are not rewarded for verifying transactions. Whilst this might seem negative from an economic standpoint as it discourages nodes from being set up, typically partners that are vested in the ecosystem are the members that run the nodes powering the blockchain. For example, the DigitalBits blockchain has several ecosystem partners running nodes on the network. Logically speaking, they are involved e.g. as application providers in the ecosystem, and thus they gain implicitly by ensuring the safety and contributing to the integrity of the network. Also, they wouldn’t act in bad faith because they would have applications running on that platform and they have the so-called “skin in the game”. Also, since it’s typically only the partners that only run the nodes, this ensures that the network is secure, whilst not wasting unnecessary energy. For reference, FBA networks like Stellar can use as much or even less energy than traditional payment systems, resulting in a more sustainable chain. The economic value of the network, as opposed to PoW, is not directly related to the energy that is being used, because DigitalBits blockchain pursues an environmental sustainability commitment since inception.

As noted by the Stellar Development Foundation, a Federated Byzantine Agreement mechanism has a range of security advantages over Proof-of-Work and Proof-of-Stake mechanisms. Broadly speaking, the Federated Byzantine Agreement Mechanism is designed to be resilient to bad actors gaining control over the network. As noted in this piece, proof-of-stake still suffers from security faults, which results in these networks having to employ measures against this, which ultimately raises energy consumption and negates some of the energy savings that Proof-of-Stake is meant to bring over Proof-of-Work.

Benefits of the Federated Byzantine Agreement

The FBA is an incredibly sophisticated consensus mechanism, so discussing it in more depth would be well beyond the scope of this piece. With that being said, however, it does have some rather sizable advantages and features. Briefly discussing them:

Transaction Speed: Because nodes can choose other nodes that they trust to assist with verifying transactions, it is very quick to verify a transaction. This is because nodes that trust each other will share the same sentiment when it comes to a transaction, making it relatively easy to meet the minimum required number of nodes verifying the transaction.
Scalability: Nodes can be added and used relatively effectively, as there is no strict requirement when it comes to adding nodes. For instance, no tokens need to be locked up to become a validator node as required by proof-of-stake or investment in expensive mining hardware as required by proof-of-work.
Transaction Count: Because transactions can occur quickly and the network scaling very well, the transaction throughput has the potential to scale rather high. DigitalBits maintains a Transactions Per Second (TPS) count of 12,000 during stress testing. For reference, many POW chains have a maximum theoretical TPS of around 400, with many large POW chains currently having a TPS of around 10. Additionally, many active POS chains have around 1000 TPS. An interesting consequence of a high TPS is that transaction fees are also much lower. DigitalBits has a transaction fee of 100 Nibbs or 0.00001 XDB.
Energy Usage: As a result of how the FBA works, very little energy is consumed by a network that uses such a mechanism, which translates to the DigitalBits network using considerably less energy than the Proof-of-Work networks which dominate the market.

The Big Picture

So to close off this report, it would be fair to address the bigger picture of what a crypto ecosystem entails. Whilst the blockchain does present the foundation upon which everything builds, it is ultimately just one element in the grand scheme of things. The partners and the technological integrations of the ecosystem all work hand in hand to build out the blockchain platform and bring value to everyone involved in the ecosystem ultimately. In the context of consensus mechanisms, there are advantages to mechanisms like the FBA for specific uses that involve big corporations, like Fortune 100, that require to comply with neutral or negative carbon footprints commitment. This brings value to the ecosystem, including partners and brands, by being secure and eco-friendly to align the interests of bringing DigitalBits blockchain to mass adoption. On the other hand, the transactions per second bring value to the clients and consumers of this ecosystem as that entails a lower price per transaction and a specific consumer focus. Thus, even though these technical topics might be of great interest, it is always important to remember that they are just a single piece in an ecosystem that should aim to provide value to any participant to fuel exponential growth.

About DigitalBits