🔦Consensus

Validation Method

The validation method is based on the BFT (Byzantine Fault Tolerance) consensus, where the result of computation achieves a majority (or unanimity, depending on the network configuration). In Lum0x, the entities responsible for this validation, commonly referred to as consensus nodes, are called operators.

However, in a diverse DeSoc environment where the types of data for consensus (catalogs) are varied, relying solely on the standard BFT method may not ensure smooth consensus. Therefore, Lum0x supports a distributed consensus algorithm that allows operators to selectively validate data catalogs, enabling Concentrated Consensus on specific catalogs.

EBFT (Economical Byzantine Fault Tolerance)

BFT is highly effective when the number of malicious nodes is low and the total number of nodes is high; otherwise, the trustworthiness of the network itself can be questioned. On the other hand, cryptoeconomic security can address these concerns but may lack in providing physical consistency and proof of consensus on computation results.

Leveraging EigenLayer’s cryptoeconomic security based on ETH PoS, we adopt an approach that combines this cryptoeconomic security with the aforementioned BFT-based algorithm.

This key principle significantly enhances our security and enables efficient consensus. EBFT is still experimental and currently operates only on the testnet. However, as EigenLayer advances and we integrate it, this approach will soon be realized.

Concentrated Consensus

If an operator chooses to validate all data catalogs, it could be seen as a standard BFT consensus responsible for agreeing on everything happening within the network (traditionally blocks). However, as previously mentioned, the complex DeSoc Data market demands real-time processing and involves various data types that are continuously generated, making this approach less practical.

Operators can now select the catalogs they wish to validate based on their economic interests. This ensures that popular and frequently used catalogs undergo more secure validation, while less popular catalogs may require additional verification procedures (Finalized time) to ensure data security.

This results in the most efficient composition for the network as a whole.