Ethereum Devs Clash Over Rising Data Costs

The Ethereum network is approaching a critical infrastructure bottleneck, sparking intense debate among developers over how to handle the blockchain's rapidly growing "state size."

At the center of the controversy is a proposed network upgrade, EIP-8037, which aims to curb data bloat by significantly increasing the upfront gas costs for developers deploying new smart contracts and storage slots.

There seems to be an economic flaw in Ethereum's current design: developers pay a one-time fee to write data to the blockchain, but network nodes are forced to pay the ongoing costs to store that data forever.

The state storage problem

Ethereum's "state" is the snapshot of all current account balances, smart contract code, and data stored on the network. Unlike transaction history, which can be archived, the state must be actively maintained and readily accessible by nodes to process new transactions.

According to network researcher @marilyn100x, the current model is unsustainable. If the network operates at a 100 million gas limit, Ethereum adds approximately 553 MiB of new permanent data daily, equating to roughly 197 GiB of new state data per year.

Currently, the Ethereum state sits at approximately 390 GiB. At the current growth rate, the network is projected to hit a critical "danger zone" of 650 GiB in less than 1.6 years. If the state becomes too large, it dramatically increases the hardware requirements to run a node, which threatens to centralize the network by pricing out average participants.

To prevent the network from hitting this limit, developers have proposed EIP-8037. Instead of time-weighted rent, the proposal acts as a deterrent by significantly raising the upfront gas costs required to create new contracts, accounts, and storage slots. Developers are incentivized to write more efficient code and avoid treating Ethereum's base layer as a cheap database.

Vitalik Buterin weighs in

The prospect of drastically higher deployment costs has prompted developers to seek alternative workarounds. On X (formerly Twitter), developer Lee Ash suggested offloading the burden to users: "What if everyone stored his own data? And the blockchain only stored the hashes? And the transactions only included the proofs?"

Ethereum co-founder Vitalik Buterin quickly shut down the idea as a near-term fix, explaining the technical limitations of cryptographic proofs in this context.

"The problem is that you need to store and update the data that the proofs are checked against, and that ends up being almost as big as the state anyway," Buterin responded.

Buterin acknowledged that alternative state management solutions exist, but he warned that they are highly complex. "There are solutions, but they have many moving parts, and all require tradeoffs relative to status quo Ethereum," he concluded.