Running an Ethereum full node might soon become significantly easier due to a proposed approach called “partial stateless nodes,” recently introduced by Ethereum co-founder Vitalik Buterin.
As the Ethereum network continues to scale, running a full Ethereum node currently requires more than two terabytes of disk storage—a substantial and growing challenge. In a recent forum post, Buterin acknowledged that while expanding the gas limit for layer-one transactions is often discussed in terms of network security, a critical problem is that higher gas limits further complicate maintaining full nodes.
Zero-knowledge (ZK) technologies, such as ZK-EVMs, offer potential paths forward by enabling users to verify the blockchain without needing a full node. However, Buterin emphasized that full nodes still play an essential role by allowing users to run local RPC servers independently, providing trustless, censorship-resistant, and privacy-focused blockchain access.
Although cryptographic techniques like Private Information Retrieval (PIR) might seem promising to increase decentralization, such solutions still tend to be costly, lack comprehensive privacy protections, and remain vulnerable to censorship pressures. Buterin expressed concern that reliance on a few dominant RPC providers could lead to increased censorship and exclusion of users from certain regions, posing broader decentralization risks to the Ethereum ecosystem.
To mitigate these issues, Buterin proposed partial stateless nodes—a novel concept whereby nodes continue verifying the entire blockchain but only store segments of the state data that are most relevant to each user’s needs. Under this model, users would select a customizable configuration to better manage node storage.
Additionally, Buterin voiced his support for completing Ethereum Improvement Proposal (EIP) 4444, which would cap the amount of historical data that full nodes are required to retain, substantially reducing overall storage requirements. He also advocated employing erasure coding techniques to enable effective and decentralized storage of older blockchain data.
This latest proposal follows Buterin’s earlier acknowledgment in May that Ethereum had become overly complex, posing obstacles to developers and slowing innovation and implementation of network upgrades. Recognizing these challenges, he had previously disclosed plans to simplify the Ethereum blockchain architecture and enhance accessibility for developers and project founders across the network.
