The ‘asynchronous’ property of Byzantine fault tolerance overcomes a challenge of fault tolerance, which is that of timing. Many forms of Byzantine fault tolerance assume there is a maximum threshold of message latency when coming to a consensus.
Asynchronous Byzantine Fault Tolerance is the highest standard of consensus algorithms. It solves the blockchain Scalability Trilemma, according to which only two of the following three components are possible at the same time:
- Decentralization
- Security
- Scalability
In an aBFT network, nodes can reach consensus independently conveys this information, and they don’t need to exchange finalized blocks. For this reason, aBFT consensus mechanisms are completely leaderless, increasing security: there is no round-robin and no Proof-of-Work.
Unlike pBFT, which relies on the fact that all the messages shared among nodes will eventually be delivered, aBFT allows for messages to be delayed or lost altogether. Besides making networks particularly resilient to DDoS attacks, aBFT also lowers the transaction’s latency, resulting in a faster network.
Finally, aBFT networks allow for greater scalability and decentralization since there isn’t excessive communication to limit the number of participating nodes.
Opera / Lachesis (by Fantom)
https://github.com/Fantom-foundation/go-opera/wiki/Lachesis
Lachesis consensus protocol which is a Directed Acyclic Graph (DAG) based asynchronous non-deterministic protocol that guarantees a pBFT consensus by creating blocks asynchronously and where the ==OPERA chain (DAG) is used for faster consensus by confirming how many nodes share the blocks.==
FANTOM A SCALABLE FRAMEWORK FOR ASYNCHRONOUS DISTRIBUTED SYSTEMS
https://arxiv.org/pdf/1810.10360.pdf
Basically, validation by the nodes in the network is not needed to arrive at consensus as occurs in typical BFT schemes.
==Fantom instead reaches probabilistic and asynchronously consensus by leveraging the concepts of distributed common knowledge==, dominator relations in graph theory and broadcast based gossip to achieve a local view with high probability of being a global view, thus increasing throughput near linearly as nodes enter the network while securing a probabilistically high enough certainty about the finality of the transactions.
==In a nutshell, Fantom achieves probabilistic consensus== which mathematically ensures high throughput and where the state of the transactions is appended to the Opera underlying chain (not a linear chain or blockchain as in Bitcoin or Ethereum but a DAG chain which makes scalability at the protocol level possible!)
This is a critical advantage that brings Fantom to different level if we consider the fact that **Lachesis will unavoidably lead to probabilistic consensus where there is no need to wait for all the nodes to reach consensus **(hence it will be fast) and once consensus is reached, then the state of the transactions will be stored in the always scalable DAG-based Opera.
https://www.youtube.com/watch?v=u0nypF5AIF4