On Thursday, Ethereum developers agreed on the full scope of the network’s changes in the upcoming ‘Dencun’ upgrade. The upgrade is expected to change many things, but one of the major changes is introducing the “proto-danksharding” feature, and it’s a really big deal, Ethereum devs say. Why, and what might be better to know about it?
Proto-danksharding introduces a new transaction type — the “blob-carrying transaction,” that is expected (EIP-4844) to allow the network to handle up to 100,000 transactions per second and drastically reduce transaction costs on Layer-2 rollups.
That “data blobs” are designed to store data for a limited time, never touching the Ethereum Virtual Machine (EVM), where all smart contracts live and interact. So blobs do not directly impact the Ethereum state or the functioning of the EVM.
“Instead, they will create a KZG commitment, which will allow us to handle data much more efficiently and in a much cheaper way, given that blobs are small enough to make disk use manageable,” says Ethereum Core developer Mario Havel.
These commitments will point to certain “blobs” of information, which is beneficial for layer-2 scaling solutions such as Optimism and Arbitrum, as it addresses the issues they currently face in accessing the necessary data for verifying transactions within their rollups. The easier it is for a rollup to access the data it needs, the faster and cheaper the transactions can be verified and reported back to the Ethereum mainnet.
The blobs set the foundation for a fully-featured data sharding. As other Ethereum Foundation members explained on the Epicenter Podcast, instead of having multiple simultaneous sidechains, there is only one chain but with “crazy” amounts of information. The data then can get split across one chain, allowing individual stakers or smaller machines to handle only a portion of the much more extensive data. These blobs will only be stored for a limited period, like a few weeks (why is it OK to delete them).
A schematic overview of blob-carrying transactions introduced by EIP-4844 (Source)
Basically, it enables the network to achieve scalability comparable with centralized computer systems. Implementing data blobs with the Dencun upgrade should make transaction data more readily available, increasing Ethereum's scalability, lowering its fees, and increasing transaction speeds on its layer-2s.
While there’s no exact date yet, the hard fork is supposed to go live by the end of 2023.
