A Low-Storage Blockchain Framework Based on Incentive Pricing Strategies

Abstract

Nowadays, blockchain bloat is an endangering issue caused by inefficient transaction storage mechanisms. Based on the Distributed File System (DFS), the blockchain network can reduce the local storage to solve the blockchain bloat problem. However, storing all blocks on DFS is not durable or scalable. Hence, classifying blocks into hot and cold was adopted in previous works. The blockchain nodes can reduce the time consumption and storage consumption by storing hot blocks locally. However, the previous works are not able to periodically check block integrity and do not provide a reward mechanism to encourage DFS system nodes to store blocks. We extend previous works based on the InterPlanetary File System (IPFS) and design an innovative scheme to incentivize IPFS nodes. The IPFS nodes are regulated with smart contracts and behave under the pricing strategy controls to increase profit. By adopting proof of retrievability, we guarantee the integrity of the blocks. Further, the redundant scheme extends our pricing strategy to improve the durability of our proposed framework. A load-balancing pricing strategy and a general pricing strategy are provided in the framework to reward the DFS nodes. Extensive experiments are presented to demonstrate that the latency and throughputs of our model are competitive, while still maintaining data integrity in the system. The additional increased throughput takes only 0.167% of that produced by the original Bitcoin and the upload latency takes only 6.67% of the mining time of the Bitcoin Mainnet. Furthermore, our load-balancing pricing strategy achieves the effectiveness to ensure the redundancy of blocks and reduces the overall storage consumption up to 97% using the load-balancing pricing strategy, compared to the non-load-balancing pricing strategy.