In LFT, consensus is based solely on message relay among participants without intermediaries. It aims to reduce the total steps required for reaching consensus as compared with preceding BFT algorithms. In traditional BTF designs, there are three steps from "pre-prepare", "prepare", to "commit", and LFT reduces these steps to 2.5 steps using a technique called spinning to simplify an overly complicated algorithm.
When the network is started, verification nodes transmit a desired transaction to the reader nodes which were determined. The primary node uses collected transactions to generate a block and sends it with a signature to the validation nodes. Validation nodes then verify the current reader has generated a block, the height of the block and the previous block hash are correct, data in the block is correct, and generates vote data which it propagates to all nodes in the network (if the first and third verification steps are correct).
Documentaries, videos and podcasts
- Byzantine Generals' problemThe Byzantine General's problem is a war scenario in which several army battalions led by different generals must collectively agree to a strategy but don't have a reliable, trustworthy way of ensuring that the other generals will all act according to the plan.
- Byzantine fault toleranceByzantine fault tolerance is a property of a distributed system such that it can tolerate components of a system failing in arbitrary ways, processing incorrect states, rather than simply stopping or crashing.
- Practical Byzantine Fault Tolerance AlgorithmPractical Byzantine Fault Tolerance algorithm (PBFT) provides Byzantine state machine replication at high speed and low cost such that it can be used for improved system safety in real world applications.
- Proof-of-work systemA proof-of-work (POW) system (or protocol, or function) is an economic measure to deter DOS attacks and other abuses (e.g. spam) on a network by requiring some work from the service requester, usually meaning processing time.