An Systematic Study on Blockchain Transaction Databases Storage and Optimization

As the peer-to-peer networks, blockchain transaction systems require each node to store a complete collection of Unspent Transaction Output (UTXO) in the in-memory transaction database, so as to ensure new transactions can by verified independently and realize the system decentralization. However, the size of blockchain transaction database is growing rapidly, in order to meet the requirements of transaction verification and system decentralization, the nodes need to constantly expand the memory capacity to adapt to the growing database, which leads to the in-memory storage problem of transaction database. This paper analyzes blockchain transaction databases and proposes a storage optimization scheme. The scheme uses the expiration identification method based on Least Recently Used (LRU) algorithm to separate the blockchain addresses into cold and hot zones. It can achieve storage optimization by moving special UTXOs away from the in-memory database, and it can achieve storage optimization without damaging data integrity and security. We present the theoretical analysis for the optimization method to validate the effectiveness. Designed extensive experiments show our proposed method outperforms the current mechanism for the blockchain transaction databases.