Ethereum Dencun Upgrade is one of the biggest updates in years to the Ethereum main net. Already established as the computer of the blockchain world Ethereum still has its problems, including inefficient data handling and therefore high gas fees. There have been some solutions such as L2s which makes transactions more efficient and cheaper yet they all have their drawbacks. The dencun upgrade will help improve layer 2s and lower their gas fees. The dencun upgrade is composed of multiple EIPs, the main one being EIP-4844.
What is EIP-4844?
EIP stands for Ethereum Improvement Proposal. This is a system where developers can suggest improvements to the Ethereum Protocol. One of the biggest problems with Ethereum is it’s high gas fees and occasionally slow processing times. EIP-4844 proposes a system of proto-danksharding which is an intial implementation of full danksharding. This is done by implementing most of the logic and scaffolding of danksharding without implementing the data fragmentation yet. This will allow for a transition that can be done over multiple network upgrades.
EIP-4844 aims to make rollups cheaper and more efficient. A rollup is a layer 2 solution that processes transactions outside the main ethereum chain but still inherit ethereum security. EIP-4844 plans on making roll ups better by introducing a new type of transaction that roll ups can leverage that will make their applications much cheaper to use.
How does EIP-4844 work?
It accepts a new type of transaction which accepts blobs (Binary Large Objects) of data to be persisted in the beacon node for a short period of time. Blobs are small enough to keep disk use manageable. Blob transactions are in the same format that they are expected to exist in future upgrades where danksharding is fully implemented (this allows for future scalability).
This comes with a blob fee market ensuring that the blob space is used efficiently and remains economically viable. This is done through blob gas as a new type of gas which is independent of normal gas.
Blobs are 4096 field elements of 32 bytes each . The blob cap per block is controlled by the MAX_BLOBS_PER_BLOCK parameter. The cap can start low and grow over multiple network upgrades. For the purpose of the Dencun upgrade the parameter is set to 6 blobs per block.
4096 field/blob * 32 bytes/ field * 6 blob / block = 0.75MB per block
This means .75MB per block of blob data is persisted in the consensus layer of the beacon node and not the execution layer. Future sharding work only requires work on the consensus layer and not the execution layer which allows for the execution layer to work in parallel on other initiatives. Blobs are short lived and only last for 18 days. This is long enough for actors of a rollup to retrieve it but short enough to keep disk usage manageable. This allows blobs to be cheaper than calldata which is data stored forever.
The crptography of EIP-4844 is based on KZG commitments which allows for efficient and secure data inclusion. This way only commitments to blobs have to be interpreted by the EVM in the execution layer and not the blobs themselves.
To make the shared secret for a KZG commitment a browser based widely distributed ceremony was so all Ethereum network participants had a chance to ensure it was generated correctly and securely. The EIP-4844 adds a new precompile called ‘point evaluation’ that verifies a KZG proof that claims that a blob evaluates to a given value at a given point.
How does this help L2s and rollups?
With this new blob space rollups will be able to put their block data into blobs rather than CALLDATA (permanent). Using the short lived block data is useful for data to be available long enough to ensure ‘honest actors’ (users) can construct the roll up space.
Quick Summary:
-Improves the scalability of Layer 2 Solutions
-Lays the ground work for further scaling into data system which implements full danksharding. (Will split ethereum network into multiple data shards each capable of storing data independently)
-Operational costs will come down because of blobs (could increase L2 and rollup usage which will bring transaction cost anyways)
Big Credit to FineMatics for the inspiration for this article. If you want a video version here is the video they did on it!
\