Computing Optimal Block Size for Blockchain based Applications with Contradictory Objectives
Nikita Singh, Manu Vardhan
Abstract
Distributed ledger technology is the driving force behind the blockchain technology and is proving its usefulness in various types of transaction processing systems. Fast, secure, reliable and efficient transactions are the key features of the blockchain based applications. A suitable or optimal size of block used by an application is dependent on the number of transactions in each block. Block size optimization is an important issue for any blockchain based application as it directly affects the performance of the application as scalability bottlenecks could prevent higher throughput and cause congestion. A larger block size will require higher transmission time compared to the smaller block size. A smaller block is more efficient but building too small a block will require higher block composition time to clear all the transactions. Both performance factors are contradictory to each other. An efficient blockchain network requires a suitable block size that demands lesser transmission time and block composition time. This paper proposes meta-heuristic algorithm based techniques for finding the suitable block size. It uses meta-heuristic algorithms to find the optimal number of transaction in each block. These algorithms are multi-objective particle swarm optimization and strength Pareto evolutionary algorithm. Experimental results reveal that the suitable block size is 213 transactions per block. Since size of a transaction is taken as 1.2 Kb, hence the results show that an optimal block size is of 255 Kbytes when network bandwidth of miners varies from 250 kbps to 1200 kbps. This shall achieve lower block transmission time and block composition time.