Propagation Delay and Mining Centralization: Modeling Stale Rates

It is explained that under certain circumstances, this strategy can provide an advantage to larger miners due to the role of block propagation times. The investigation is grounded in a simplified model which seeks to elucidate how these propagation times influence a miner's revenue alongside their share of the network's overall hashrate. Assuming a uniform propagation time and instant block publication upon discovery, the model outlines two key scenarios: one where a competing miner finds a block first, leading others to mine on that block, and another where a block discovered shortly after increases the likelihood of the original block becoming stale, particularly if a subsequent block is quickly found. The probability of a block turning stale is heavily influenced by the miner’s control over the network's hashrate, with centralization emerging as a significant factor. Larger mining entities or pools are more likely to win in these scenarios, suggesting that block propagation times can inadvertently favor mining centralization.

The study underscores the complex interplay between block propagation times, miner strategies, and the broader implications for the fairness and decentralization of the mining process. By illustrating how slight variances in propagation times can markedly affect revenue, the analysis points to the importance of careful pool selection based on size and the potential impacts of P2P network modifications on mining dynamics. The discussion suggests that advancements in the P2P network that reduce propagation delays could render mining more equitable among differently sized operations, thereby addressing some aspects of centralization. This assertion is bolstered by data visualizations and simulation results, accessible via links such as the detailed analysis of Selfish Mining and the source code for these simulations on GitHub.

Additionally, the narrative touches on the misconceptions about average revenue models within mining operations, highlighting the substantial portion of revenues often consumed by power costs, thus overstating miners' net profits. This point emphasizes the complexity of accurately evaluating financial outcomes for miners, suggesting a reality far more nuanced than simple revenue calculations would imply. Further insights are offered into the management of stale blocks and the visibility within the network, particularly following the implementation of compact blocks. A proposed solution includes introducing a new "STALEHEADER" message protocol to enhance network efficiency by enabling the relay of headers closely connected to the best chain tip without necessitating the ability to serve the entire block. This mechanism aims to improve statistics on network performance by tracking stale headers near the chain's tip or those accepted as stale, thereby increasing visibility into large invalid forks while maintaining protection against DOS attacks. This proposal reflects ongoing efforts to refine blockchain protocols, addressing operational challenges while ensuring network integrity, as discussed among Bitcoin developers. For further exploration on benchmarking, the link provided is https://stratumprotocol.org/blog/hashlabs.