Ethereum co-founder Vitalik Buterin has dismissed a proposed fix for the network’s swelling state size, warning that relying on users to store their own data while keeping only Merkle roots on-chain would not solve the fundamental problem. The rejection comes as developers debate EIP-8037, a proposal to increase upfront costs for creating smart contracts and storage slots, aiming to curb permanent data accumulation.
Ethereum’s state — encompassing wallet balances, smart contract code, and account data that nodes must retain — is expanding at a rate of roughly 553 MiB daily under a 100 million gas limit, according to researcher marilyn100x.eth. That translates to nearly 197 GiB of new state data per year, with the current total near 390 GiB. Researchers warn that at this pace, the state could hit a critical 650 GiB threshold in under two years, raising hardware demands that could threaten node decentralization.
In response, EIP-8037 seeks to make permanent storage more expensive at the time of deployment, in contrast to the current one-time fee model that leaves node operators carrying the burden indefinitely. Buterin addressed a user-stored-data alternative, where only cryptographic hashes reside on-chain and users keep the full record. He explained that the proof verification data required by such a system would still need to be stored and updated, and its size could eventually rival the full state, negating efficiency gains.
“Solutions exist, but they require significant trade-offs and complex components compared to Ethereum’s current structure,” Buterin wrote. His comments underscore the tension between affordability, security, and decentralization as the community weighs EIP-8037 and other long-term state management strategies. For users, unaddressed state bloat could drive up gas fees and squeeze out smaller node operators; for developers, the debate highlights the need to design storage-efficient applications from the outset.
