Ethereum has challenges in scalability since its inception, and EIP-4844 is one of the solutions to address this issue through the concept of sharding.
What is sharding
Sharding basically is a way of dividing up the work of processing transactions so that it can be done more efficiently. Instead of having one big computer processing all transactions, the work is split up among many smaller computers, each handling its own piece of the puzzle.
Ethereum sharding isn't just about processing transactions faster, it also provides more space for data. This means that the Ethereum network can handle more complex data and support high-throughput transactions.
EIP-4844 is the first iteration of the sharding design of Ethereum, which is a way for Ethereum to have more data go through the network.
Data availability on the Ethereum network is one of the primary scaling bottlenecks. To address this issue, EIP-4844 is trying to add data availability capacity on Ethereum without breaking composability and the execution layer on L1.
Instead of interpreting the data in a transaction, Ethereum sharding verifies that the data is available for download from the network. This means that the Ethereum protocol doesn't have to interpret the data, making the transaction processing faster and more efficient.
The above all sounds cool - but it is quite a long time away.
In the meantime, Proto-danksharding, also known as EIP-4844, is a proposal to improve the scalability of Ethereum without implementing sharding yet.
It introduces a new type of transaction called a blob-carrying transaction, which is similar to regular transactions but also carries a large piece of data called a blob.
The blob data is not accessible to EVM execution, but it can be much cheaper than similar amounts of calldata. Although validators and clients still have to download the full blob contents, data bandwidth is limited to 1 MB per slot, which increases scalability gains without competing with existing Ethereum transactions.
EIP4844 reduces transaction fees on L2s
The main benefit of EIP-4844 is that it creates a cheap place for L2 solutions to post data on Ethereum and reduces overall transaction fees users pay on L2 significantly; directly impacting users as they will directly benefit as they will be paying fewer fees on their favorite L2.
The blobs data does not have to be available forever, but for a certain period (e.g. 1-3 months), long enough that allows L2s to ensure at least one honest actor to reconstruct the state and challenge or replace the bad sequencer.
EIP-4844 provides significant scaling relief for rollups by allowing them to initially scale to 0.25 MB per slot, with a separate fee market for blobs allowing fees to be very low - almost up to 10x cheaper.
In summary, EIP-4844 is a solution to address Ethereum's scalability issues by providing data availability capacity on Ethereum without breaking composability and the execution layer on L1.
It reduces overall transaction fees users pay on L2 and provides significant scaling relief for rollups. This is an exciting development for Ethereum, as it allows the platform to handle more data while maintaining its decentralization and security features.
It is expected to go-live in 2H 2023 so stay tuned.