Unpacking Long-Latency Transactions in Ethereum

Abstract

Blockchain systems such as Bitcoin and Ethereum have limitations in efficiency, resulting in an inability to immediately confirm all transactions, leading to extended periods of transactions residing in the mempool. We refer to these transactions as “long latency trans- actions” and this paper explores the issue of resource utilization in- efficiencies issues from these transactions. Utilizing the Geth client, the study quantifies the impact of these transactions on Ethereum’s resource consumption, which encompassing three crucial metrics: computational power, memory storage, and network bandwidth. Furthermore, this study also identifies three primary factors con- tributing to long latency transactions: low gas prices, long block processing times, and future-index transactions. Through empirical analysis, this study offers insights into the transaction-handling mechanisms in Ethereum. The implications of our findings aim to contribute to the enhancement of resource efficiency within the Ethereum blockchain ecosystem.