Efficient mempool synchronization using bloom filters and entropy coding

One of the major challenge to scale bitcoin is that a lot of the work done by nodes is reactive. Nodes receive a block and start validating. This takes time and resources and is right in the hot path. The problem gets worse the bigger the block are, to the point were consensus takes more and more time to be reached. At some point, consensus cannot be reached.

In order to improve scaling capabilities of the network, it is key that nodes do as much work as possible ahead of time. When a block is received, node just have to assert that the node effectively contains what they expected, which speeds up validation significantly.

Continue reading