WiMi is Working on a New Mixed Byzantine Fault Tolerant (MBFT) Algorithm to Enhance Blockchain Fault Tolerance

Beijing, Jan. 16, 2024 (GLOBE NEWSWIRE) -- WiMi Hologram Cloud Inc. (NASDAQ: WIMI) ("WiMi" or the "Company"), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced that it is working on a novel mixed byzantine fault tolerance (MBFT).

MBFT has its origins in the PBFT algorithm, the first consensus algorithm used in the Union Blockchain, but in many ways, MBFT has improved and innovated. MBFT improves scalability while ensuring fault tolerance. It combines sharding and layering to improve the performance of the blockchain. Layering effectively separates the functions of nodes, and by assigning the entire verification function and demodulation process to different nodes, layering reduces the load on a single node and improves consistency efficiency. Sharding can assign transactions to different groups of nodes, and when the number of transactions increases significantly, the system can dynamically increase the number of nodes and shards, thus improving processing capacity and reducing latency. Through functional node division, MBFT improves the efficiency of the blockchain without sacrificing security. In addition, MBFT introduces a random node selection mechanism and a credit mechanism to further enhance its security and fault tolerance. In MBFT, we assign nodes to different consensus groups and design new transaction verification and fault tolerance algorithms for each consensus group. Each consensus group verifies only a part of the transaction. At the same time, the blockchain system can still maintain 1/3 fault tolerance. MBFT algorithm can reduce blockchain forks while ensuring high fault tolerance and speed.

MBFT's two-layer design and sharding greatly reduce the amount of communication between nodes, resulting in higher efficiency. To select consensus nodes, MBFT designs a randomized node selection algorithm based on VRF and Feldman VSS. This design allows the selection phase to overlap with the normal block-packing phase in the normal operation of the system without additional time. Meanwhile, a global reselection mechanism and a credit evaluation mechanism are designed in combination with the characteristics of controllable verification nodes in the federated blockchain.