Author: Bob McElrath 2016-05-11 20:06:48
Published on: 2016-05-11T20:06:48+00:00
The email discusses the use of Bloom filters and Cuckoo filters in Bitcoin to prevent light clients from downloading more blocks than necessary. The author suggests that using a fixed false-positive rate for expected wallet sizes makes more sense and is faster, and that the false-positive rate should be less than 1/block height to minimize unnecessary downloads. The required size of the filter commitment is roughly N log2 H, where H is the block height. If Bitcoin had these filters from the beginning, a light client today would have to download about 12MB of data in filters, while personal SPV wallets currently use 31MB. The author concludes that it's not clear whether this approach is a bandwidth win, but it's definitely a win for computing load on full nodes. A link to a paper on Cuckoo filters is provided.
Updated on: 2023-06-11T04:54:47.093434+00:00