The Dencun upgrade was an Ethereum mainnet hard fork activated on March 13, 2024. It combined Cancun, the execution-layer upgrade, and Deneb, the consensus-layer upgrade. Its principal scaling contribution was introducing a dedicated way for layer-2 rollups to publish transaction data to Ethereum at potentially lower cost. Dencun is a completed protocol upgrade, not an upcoming release.
How Dencun Changed Rollup Data Publication
Layer-2 rollups execute transactions outside Ethereum mainnet and collect them into batches. They publish data to Ethereum so their activity can be reconstructed and checked, rather than requiring Ethereum's execution layer to process each transaction individually. This separates rollup execution from the task of making the underlying data available.
Before the Dencun upgrade, rollups commonly published this data as calldata: bytes supplied with ordinary Ethereum transactions. Calldata is part of execution-layer transaction history and is priced through the normal gas market. Publishing large batches this way can be a significant rollup expense.
Dencun included EIP-4844, also called proto-danksharding, which introduced blob-carrying transactions. EIP-4844 was one proposal within the wider Cancun–Deneb package, not a separate upgrade combined with it.
How Blobs Work
A blob is a dedicated data payload attached to a transaction, designed to provide temporary data availability. Ethereum handles blob contents through its consensus layer rather than storing them permanently as execution-layer calldata. Smart contracts cannot read blob contents directly in the same way they can read calldata.
Under the Dencun upgrade, a rollup can publish its batch data in blobs while using Ethereum transactions for the parts of its workflow that require on-chain execution. Ethereum nodes retain and serve blob data for a limited period, allowing it to be retrieved and checked. Blobs are therefore not generic off-chain file storage or a permanent archive. Systems needing a longer history must arrange separate retention; temporary blob retention does not mean that rollup balances or transaction outcomes expire with the data.
Looking for liquidity, exploring on-ramp/off-ramp services, or seeking expert guidance?
Get startedWhich Fees Dencun Can Reduce
Blobs have a separate fee market from ordinary execution gas. Their price responds to demand for blob capacity, so rollups no longer need to use calldata for every byte of batch data they publish. Blob-carrying transactions still have execution-layer gas costs; the separate market applies to the blob data.
The main fee benefit of the Dencun upgrade is lower rollup data-publication costs, which can support lower layer-2 user fees. The amount saved depends on whether a rollup uses blobs, demand for blob space, and how its fee policy passes costs on to users. A rollup's user fee can also include execution and other operating costs. Blob fees can rise when demand increases, so neither fixed savings nor uniformly cheap transactions are guaranteed.
Scope and Limitations
The Dencun upgrade did not guarantee cheaper transactions on Ethereum mainnet. A mainnet token transfer or smart-contract interaction still uses execution gas, whose price depends on demand for execution capacity. Nor did the introduction of blobs make transaction execution intrinsically faster.
Proto-danksharding was a step toward further data-availability scaling, not full danksharding. Its role was to introduce the blob-based foundation rather than deliver every planned scaling feature. As a hard fork, Dencun required node and validator operators to use compatible client software. Ordinary ETH holders did not need to convert their ETH or exchange it for a new token.
