Solana proposal would eliminate computational bottlenecks
Solana (SOL) developers proposed a protocol change that would profoundly impact its ecosystem.
The proposal in question would introduce a “lattice-based” hashing protocol that alters how the blockchain tracks and verifies balances and on-chain data in a very significant way. The proposal’s GitHub page reads:
“The main goal is to scale Solana to billions of accounts, and compute a “hash of all accounts” in practical time and space.”
The detailsIf approved, the proposal would introduce a new blockchain scalability approach that could be applied to other protocols. Solana and other blockchains regularly recalculate the entire state of all user accounts, which requires ever-increasing resources as the network grows.
In the context of blockchain, “state” refers to the current snapshot of all data on the network — such as account balances, smart contract storage, and other information— after all validated transactions have been processed. It’s essentially the shared, up-to-date database that every node agrees upon at any given block.
Experts often call This issue the “state growth problem” and the proposal addresses this exact concern. Solana Labs co-founder Anatoly Yakovenko explained in late May 2024:
“Each day, about [one million] new accounts are added to the chain. The total state right now is over 500 [million], while the snapshot size is about 70GB. These numbers on their own are totally manageable as hardware improves, but the goal of the [Solana’s smart contract processor] is to provide the cheapest possible way to access the hardware, and to achieve that the state and memory have to be managed within current hardware constraints.”
Yakovenko also explained that “new account creation has to actually create new accounts” and, as a consequence, “a new account has to prove that it is new somehow.” If the proposal is implemented this requirement would be eliminated by enabling instant verification with no need to recalculate the state.
The post Solana proposal would eliminate computational bottlenecks appeared first on ReadWrite.