Research And Application Of Blockchain Scalability Based On PBFT Algorithm

As an emerging technology in the field of distributed consensus,blockchain has attracted extensive research and attention in recent years.However,the current blockchain system still faces the problem of scalability.To solve this problem,researchers have proposed a variety of solutions,such as Side chain,Cross chain,Fragmentation technology and so on.Fragmentation technology is considered to be the best solution to solve the scalability of blockchain,but there are still some problems in Fragmentation technology.One is that when nodes are allocated,if the node partition is not reasonable,the probability of Byzantine nodes allocated to the same partition will be increased,and the credibility of fragmentation data can not be guaranteed.The other is that there are many signature information in the block,Block transmission and signature verification are inefficient.In order to solve the above problems,on the basis of PBFT algorithm,this thesis proposes a node random allocation algorithm and an improved APBFT algorithm based on PBFT algorithm,and applies the improved algorithm to the credit mutual recognition system of MOOC.The specific research contents are as follows:(1)In the process of node partition,if the node is biased to a certain partition allocation or the node can control the choice of partition at will,it may cause Byzantine nodes to collude in the partition,which can not guarantee the credibility of single partition data.Therefore,it is very important to realize the unpredictability and randomness of node allocation.In order to solve this problem,a node random allocation algorithm is proposed.This algorithm mainly allocates nodes randomly through consistent hash algorithm.However,in consistent hash algorithm,the weight of all nodes is the same,and there is no corresponding load balancing strategy,Therefore,the weight of nodes is used to optimize the consistent hash algorithm.(2)When the nodes in the partition reach a consensus through PBFT algorithm,they need to sign the block and store the signature information in the block,which will result in a large amount of block data and low efficiency of block transmission and signature verification.In order to solve this problem,on the basis of PBFT algorithm,an improved aggregate practical Byzantine fault tolerance(APBFT)algorithm based on aggregate signature is proposed.By compressing multiple signatures with aggregate signature,the block transmission pressure and signature verification time are effectively reduced,the consensus time is shortened,and the scalability of the system is improved.(3)Online education represented by MOOCS has developed rapidly in recent years.In order to solve the problem of lack of credibility in MOOCS credit recognition,the improved APBFT algorithm is applied to MOOCS credit recognition system,and the system is designed and implemented.