Currently, all blockchain consensus protocols (eg. Bitcoin, Ethereum, Ripple, Tendermint) have a challenging limitation: every fully participating node in the network must process every transaction. Recall that blockchains have one inherent critical characteristic — “decentralization” — which means that every single node on the network processes every transaction and maintains a copy of the entire state. While a decentralization consensus mechanism offers some critical benefits, such as fault tolerance, a strong guarantee of security, political neutrality, and authenticity, it comes at the cost of scalability. The number of transactions the blockchain can process can never exceed that of a single node that is participating in the network. In fact, the blockchain actually gets weaker as more nodes are added to its network because of the inter-node latency that logarithmically increases with every additional node. In a traditional database system, the solution to scalability is to add more servers (i.e. compute power) to handle the added transactions. In the decentralized blockchain world where every node needs to process and validate every transaction, it would require us to add more compute to every node for the network to get faster. Having no control over every public node in the network leaves us in a pickle. As a result, all public blockchain consensus protocols that operate in such a decentralized manner make the tradeoff between low transaction throughput and high degree of centralization. In other words, as the size of the blockchain grows, the requirements for storage, bandwidth, and compute power required by fully participating in the network increases. At some point, it becomes unwieldy enough that it’s only feasible for a few nodes to process a block — leading to the risk of centralization. In order to scale, the blockchain protocol must figure out a mechanism to limit the number of participating nodes needed to validate each transaction, without losing the network’s trust that each transaction is valid. It might sound simple in words, but is technologically very difficult. Why?