# End-to-End Latency Analysis and Optimal Block Size of Proof-of-Work Blockchain Applications

arXiv: https://arxiv.org/abs/2202.01497 · HTML: https://mlsanity.com/paper/2202.01497/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Feb 2022.

## Built on by (most cited first, 3 of 3)

- [The Cost of Training Machine Learning Models over Distributed Data Sources](https://mlsanity.com/paper/2209.07124/) (2022)
- [On the Decentralization of Blockchain-enabled Asynchronous Federated Learning](https://mlsanity.com/paper/2205.10201/) (2022)
- [The Implications of Decentralization in Blockchained Federated Learning: Evaluating the Impact of Model Staleness and Inconsistencies](https://mlsanity.com/paper/2310.07471/) (2023)

## Abstract (arXiv)

Due to the increasing interest in blockchain technology for fostering secure, auditable, decentralized applications, a set of challenges associated with this technology need to be addressed. In this letter, we focus on the delay associated with Proof-of-Work (PoW)-based blockchain networks, whereby participants validate the new information to be appended to a distributed ledger via consensus to confirm transactions. We propose a novel end-to-end latency model based on batch-service queuing theory that characterizes timers and forks for the first time. Furthermore, we derive an estimation of optimum block size analytically. Endorsed by simulation results, we show that the optimal block size approximation is a consistent method that leads to close-to-optimal performance by significantly reducing the overheads associated with blockchain applications.