Author: Andrew Poelstra 2017-04-20 20:32:12
Published on: 2017-04-20T20:32:12+00:00
In this email exchange on bitcoin-dev mailing list, Andrew Poelstra responds to a proposal made by Tom Zander in which he suggests pruning the blockchain by randomly storing only 25% of all blocks. David Vorick had previously criticized this idea, highlighting the need to connect to many nodes to download the entire blockchain and the fact that the probability of one block being dropped by everyone increases significantly with this method. Poelstra supports Vorick's argument, stating that even if there are no archival nodes in the network and all nodes are storing only one random index, the chance of five arbitrary nodes giving unique indices is around 98.4%. In contrast, with the proposed 25% pruning scheme, it would be necessary to connect to almost 60 nodes to obtain the same level of probability, leading to a ten-fold increase in the number of nodes required to download the whole chain. Moreover, using this method could lead to potential denial of service attacks due to the assumption that not all nodes will adopt the new mode of pruning.
Updated on: 2023-06-12T00:22:29.532044+00:00