Raft consensus method suitable for Internet of Things

Abstract

The invention discloses a Raft consensus method suitable for the Internet of Things, and the method comprises the steps: collecting the Internet of Things data in a distributed physical network in real time through an Internet of Things collection device, and caching the Internet of Things data in an Internet of Things gateway; packaging the Internet of Things data within the time of 60s + K / N, and counting the packaged data traffic; calculating a random number Non value based on a random number function of a linear congruence algorithm and a Hash algorithm; according to the data flow parameters, carrying out master selection difficulty dynamic adjustment, determining a target value, and judging whether the random number None value accords with the target value or not; designing a timer, starting a consensus main selection process, selecting a consensus main node according to a random number Nonce value conforming to a target value, and performing RAFT log copying. The dynamic election of the main node is realized according to the dynamic change of the data volume received by the node receiving and collecting equipment, and the method is safe, reliable and low in operation and maintenance cost.

Description

technical field [0001] The present invention relates to the technical field of consensus mechanism, in particular to a Raft consensus method suitable for the Internet of Things. Background technique [0002] The RAFT consensus algorithm is an algorithm for realizing distributed consensus, which is mainly used to manage the consistency of log replication. This consensus algorithm is adopted by many distributed systems because of its easier understanding and high consensus efficiency. [0003] However, since the traditional RAFT consensus algorithm uses multiple communication voting methods, this puts forward higher requirements for the communication efficiency of distributed systems that usually use P2P networks. At the same time, the RAFT consensus algorithm based on voting is very random. , although a certain degree of security is guaranteed, but the degree of centralization is too high to be applicable to a weak trust environment. The high requirements of the communicatio...

AI Technical Summary

Problems solved by technology

[0003] However, since the traditional RAFT consensus algorithm uses multiple communication voting methods, this puts forward higher requirements for the communication efficiency of distributed systems that usually use P2P networks. At the same time, the RAFT consensus algorithm based on voting is very random. , although a certain degree of security is guaranteed, but the degree of centralization is too high to be applicable to a weak trust environment

The communication environment requires too much and is not suitable for the weak trust environment, so that the traditional RAFT consensus algorithm cannot be coupled with the application scenario of the Internet of Things

Images