1. Overview of MonadBFT

The MonadBFT consensus mechanism is a method that ensures the nodes in a blockchain network can reach consensus on the network’s state, even in the presence of some faulty or malicious nodes (Byzantine Fault Tolerance – BFT). To provide a clearer explanation, let’s go through the basic concepts and how MonadBFT operates.

A BFT system must ensure:

• Consistency: Honest nodes must agree on a single result.
• Safety: Once a block is confirmed, no node can change this decision.
• Liveness: Even if some nodes fail or act maliciously, the system can continue to operate.

2. How MonadBFT Works

MonadBFT is a variation of modern BFT consensus mechanisms, drawing inspiration from popular BFT solutions such as HotStuff and PBFT. It operates in rounds and features key differences from previous mechanisms.

The MonadBFT consensus mechanism ensures that nodes in a blockchain network can agree on the network’s state, even when some nodes are faulty or malicious (Byzantine Fault Tolerance – BFT). Let’s dive into its core concepts and operational flow.

3. Round-Based Operation

MonadBFT works in rounds, and for each round, a randomly selected leader proposes a new block. The process follows these steps:

• Round K:
  • A leader is randomly chosen to propose a block.
  • Validators receive this proposal and vote on it.
  • If more than 2/3 of the validators agree the proposal is valid, a Quorum Certificate (QC) is created, which serves as proof that the block has been approved by a majority of the nodes.
  • If there aren’t enough validators agreeing or there’s an issue with the proposal, validators issue a Timed-out Certificate (TC), signaling the round’s failure.
• Round K+1:
  • A new leader starts this round and may receive a QC or TC from the previous round.
  • If the leader receives the QC from round K, they use it to create a new block proposal and continue the process.
• Round K+2:
  • When a QC from round K+1 is received, the proposal from round K is formally confirmed, meaning that the block from round K is added to the blockchain.

4. Quorum Certificate (QC) and Timed-out Certificate (TC)

• QC (Quorum Certificate): A certificate generated when more than 2/3 of validators agree on the leader’s block proposal, proving the block was approved by most of the honest nodes.
• TC (Timed-out Certificate): If consensus isn’t reached due to timeout or other issues, a TC is created to indicate the failure of the round.

5. Pipeline Mode of Operation

MonadBFT employs a pipeline-like mechanism where block confirmation happens continuously, with each round handling part of the confirmation process. This continuous processing increases the speed and efficiency of consensus.

6. Differences from Traditional BFT Mechanisms

MonadBFT is inspired by the HotStuff protocol (a modern BFT protocol) but comes with several improvements:

• Reduced Latency: While HotStuff has a three-phase latency, MonadBFT eliminates this, speeding up consensus.
• Low Communication Overhead in Normal Conditions: When the leader performs well, communication costs are linear in MonadBFT. However, when a leader change occurs, costs may rise to quadratic, ensuring the system can still achieve consensus in two steps.

7. Chaining Blocks

In MonadBFT, blocks are chained together, meaning that each round not only processes a block but also ensures that the block from the previous round is confirmed. This continuous chaining ensures smooth and efficient block confirmation.

8. Example

Imagine a MonadBFT blockchain network with 10 nodes, where one node is randomly chosen as the leader for round K. This leader proposes a block and sends it to the validators:

• If 7/10 validators (more than 2/3) agree that the block is valid, a QC is generated, and the next leader in round K+1 continues the process.
• If not enough validators agree, or an error occurs, validators create a TC, signaling that the round has failed and needs to move to the next one.

9. Conclusion

MonadBFT is an advanced consensus mechanism based on the BFT model, ensuring fast and secure consensus even when some nodes are faulty or malicious. It uses a pipeline approach for continuous block processing and improves consensus efficiency by reducing latency and communication overhead under normal conditions.

With MonadBFT’s efficient consensus mechanism, the platform establishes stability and speed in transaction confirmations. But how does Monad optimize the processing of batches of transactions simultaneously? That’s where Deferred Execution, which separates consensus and execution, plays a groundbreaking role. Let’s explore this mechanism further in the next section: <<Deferred Execution – Monad Technical Overview>>

READ MORE

🔹MonadBFT: A Consensus Mechanism Delivering High Performance for the Monad Blockchain

🔹Deferred Execution: Optimizing Consensus and Parallel Transaction Processing

🔹Parallel Execution: Enhancing Multithreaded Transaction Throughput

🔹MonadDB: A Data Layer Optimized for Parallel Execution