Author: Christian Decker 2011-12-21 16:10:45
Published on: 2011-12-21T16:10:45+00:00
The discussion revolves around the emergence of supernodes that can validate transactions and store them for miners in a hierarchical network. These nodes would be owned by miners themselves, who can specialize in verifying all transactions included in the block or have someone else do it. The network would have miners/supernodes connected tightly at the top and lightweight clients verifying only transactions destined for them at the bottom. A brainstorming session was held on the forum to discuss distributed hash tables (DHTs) and their role in reducing the risk of a sybil attack. While some participants suggested using IP addresses, others were wary of including the port as it could pose a security risk. Broadcasting transactions may not be necessary since miners can fetch them from storage places. Moreover, internal routing within the DHT could lead to caching at nodes leading to the actual location, preventing flashcrowds and reducing load on responsible nodes. Michael Grønager suggested the possibility of using browser-based wallets that can act as supernodes. However, if bitcoin wants to maintain a real peer-to-peer setup and enable scalability, they need to follow the proposed path. He also discussed the potential new attack vector and suggested dividing the hash space no more finely than when you can find eight different A.B nodes at all times. Eric Lombrozo, in response, discusses the inevitability of supernodes emerging that can validate transactions for other nodes and partition them into hashspace. Anonymous peers have no reputation to defend while trusted supernodes would, providing an incentive to ensure legit nodes. However, unless the supernode validates the entire block chain and transaction pool itself, it could only assess trustworthiness through random sampling.
Updated on: 2023-06-05T00:45:52.366516+00:00