I've been thinking about the SPV implementation... I assume that for V13 we'll just piggyback of electrum-server and use that.
I've been thinking about the efficiency and usefulness of doing it in the core dashd process... since we're already using leveldb for storage, and since SPV is pretty much requisite for most real-world use-cases of FinTech (e.g. merchant solutions). I've been considering consulting a specific expert on algorithms to see if a burst tree (trie?) would be a more efficient option for UTXO storage/retrieval than a Patricia tree, which is what I believe electrum-server uses.
Anyone have input/ideas on this?
I've been thinking about the efficiency and usefulness of doing it in the core dashd process... since we're already using leveldb for storage, and since SPV is pretty much requisite for most real-world use-cases of FinTech (e.g. merchant solutions). I've been considering consulting a specific expert on algorithms to see if a burst tree (trie?) would be a more efficient option for UTXO storage/retrieval than a Patricia tree, which is what I believe electrum-server uses.
Anyone have input/ideas on this?