October 28, 2022

DankSharding - What is it and how does it work?

Blockchain community, have you heard about sharding? Well, there is a concept that exists above this one, and it is called Danksharding.  

As we know, the Ethereum community has increased a lot, and the community keeps researching cryptography, consensus protocols, computer science, etc. 

In this case, the scale in Ethereum is primordial for the community that will utilize sharding and then, to execute this concept, execute Danksharding.

What is Danksharding

Danksharding is the new concept for sharding used in Ethereum. It was created by Dankrad Feist, a veteran researcher of Ethereum, to avoid the MEV (Maximal Extractable Value), increasing decentralization and security. 

It is also known as EIP-4844 and appears on the EIP list of Ethereum. Two roles exist: constructor and proposer, the first one are the users that get the list of transactions from the proposers. 

The constructor car re-orders transactions to maximize MEV in terms of constructor benefits, but also for the users that execute transactions.

Traditional Sharding v Danksharding
Traditional Sharding v Danksharding - Source

As we can see in the above image, the blocks of execution and sharding build together, which gives all the validation data to execute concurrently. And as a result, there is less time to confirm the shard blocks, and all are visible for layer 1 all the confirmations. 

Also, Danksharding will allow calls between ZK-Roll Ups (Arbitrum, Optimism) and Layer1, increasing the scalability and the speed per transaction in the mainnet. 

In Danksharding, there are no different block proposers; there is only one, and with Danksharding, nobody knows the content on the mempool. 

DankSharding Solutions

The solution of DankSharding uses PBS (Proposer-Builder Separation) and crList. The first concept means the unique element of actors called builders that bid for the right to vote slot content.

There is an implementation in the Ethereum repository on GitHub. This idea was real for Dankrad Feist, one of the most prominent and veteran researchers from Ethereum. 

But, right now, there is a big problem with Danksharding: it uses too much memory. Feist calculates 2,5 TB per year, more memory than Ethereum uses now after the merge - a change to all the consensus protocols of Ethereum PoW (Proof of work) to PoS (Proof of Stake). 

Changes after the Merge

After the merge, there were some significant changes. It is estimated that, including the complete shards, there will be around 40 TB of historical data of blobs per year. As a result, ongoing work needs to be done with the size of memory. 

This graphic shows the differences in Gas used per block and the Basefee:

Gas used per block on BaseFee
Gas used per block on BaseFee - Source

This re-design of the sharding improves the usability of Arbitrum and the oracles chainlink and increases the way to use L2. 

And right now, the Ethereum community is working to make it a reality and has questions like the Exact details of the trusted setup implementation and if blobs are priced in Gas or ETH.

It would be ready in 2023 when sharding is applied that year, but other parts of the community also say it would be possible to implement it in 2024.

Conclusion on DankSharding

There is much work to do, but DankSharding provides confidence in the scalability solution of Ethereum and real and sophisticated solutions to increase the speed and the performance of the execution of the transactions, including MEV advantages and disadvantages.

What are your thoughts on DankSharding? Do you share this confidence? Let us know in the comments section.