We’re rewarding the question askers & reputations are being recalculated! Read more.

Hot answers tagged

4

This paper Performance Evaluation of the Quorum Blockchain Platform gives a good comparison between Raft and IBFT in term of latency and throughput.


1

Quorum's raft implementation is specifically designed for the speed of txn processing and confirmation, thus Raft mode is going to be much faster than IBFT.


1

The equation to solve that problem is the following: The system can tolerate at most of F faulty nodes in a N validator nodes network, where N = 3F + 1 You can visit this link https://github.com/ethereum/EIPs/issues/650 in Consensus part to more information.


1

A node that is not party to a private transaction can still mine (mint) the transaction. The minting node is selected by whichever consensus mechanism you have configured. A simplified description of how private transactions work is as follows: Node A sends the transaction to its local transaction manager and is returned a hash to the transaction. The ...


1

Minimum for Raft is 3 nodes and minimum for Istanbul is 5 validator nodes.


1

To reach consensus w IBFT in the presence of F s faulty nodes require N > 3F validator nodes, or N=3F+1. So to Nathan's original question, the total nodes are the former. To progress in the 3-phase commit requires responses from 2F+1 nodes; so you need ~ 2/3 of the nodes to respond appropriately to reach consensus (7 out of 10 in the case F=3).


1

IBFT mode of Quorum is 2F+1. Heres the actual EIP with detailed explanation of the IBFT rounds: https://github.com/ethereum/EIPs/issues/650


Only top voted, non community-wiki answers of a minimum length are eligible