Not long ago, Vitalik, the founder of Ethereum, said at the ETH Shanghai Summit that the merger of Ethereum may begin in August, and the first major event after the merger is EIP-4844, proto-danksharding, and proto-danksharding is only the first of Danksharding. step.
On a recent Bankless podcast, merger coordinators Tim Beiko and Vitalik Buterin, and Ethereum researchers Dankrad Feist and Protolambda had an in-depth conversation on the topic of proto-danksharding and Danksharding.
The following is the main content of the conversation.
What is Danksharding
- Danksharding and proto-danksharding are iterations on Ethereum’s sharding design
- Sharding is a way for the Ethereum network to process more data. “All of these sharding approaches have the same end goal of creating a cheap place for second-layer solutions to publish data on Ethereum.” Tim Beiko
- EIP-4844 是 proto-danksharding
- Proto-danksharding is the first step towards sharding
- Danksharding is a simplification of the previous roadmap
Moving from full execution sharding to data-only sharding
- The sharding roadmap has been continuously simplified since its release in 2016
- The first step is to decide not to bother with anything other than quadratic sharding
- The second step is to have each shard block directly included in the beacon chain
- Shard blocks no longer contain transactions that will be executed at the Ethereum layer, but instead shard blocks contain only large blocks of data
- The layer 2 rollup protocol will be responsible for creating a secure and scalable experience for its users using this data space
- Next is Danksharding, where only one proposer chooses all shard blocks and shards that appear in a particular beacon block
- Full sharding currently halfway through the timeline
Why can Danksharding be achieved without sacrificing the decentralization of the chain?
- Proof-of-Stake has proposer-builder separation
- Traditionally, block proposers build blocks
- Proposal is simple and cheap, while building is a complex process
- Build entities better suited for capitalization – people with large machines that can process large amounts of data
- Anyone can verify that the block is correct
How does EIP-4844 help with Layer 2?
- Danksharding introduced data availability sampling. It tries to better distribute jobs over the network
- They can be optimized for the security properties required by L2
- How it helps tier 2: publish data and ensure that the honest few securing L2 have access to data first; ensure participants have access to data even in the event of downtime, censorship, unforeseen circumstances, etc.
What is data availability and how it differs from data stored on Ethereum
- Data availability: Does the data go through a publishing process, broadcast on the public network, and anyone who wants to download the data has time to do so?
- Ethereum is able to provide consensus on data availability when comparing Ethereum to IPFS
- Layer 2 protocol depends on the data available there
- ZK Rollup example: Sequencer accepts transactions —> sequencer publishes blocks containing state-delta; sequencer manages internal state.
- The difference between ZK Rollup and Validium:
- In ZK Rollup, state-delta is on-chain
- In Validium, only proofs are on-chain. Everything else is off-chain
- The only difference is what happens when the sequencer disappears. In Validium, this is a problem because if the sequencer disappears, the funds are stuck there forever
- The beacon chain will only contain the hash of the data
- Shards do not need to store data forever. The key is to give people who want to download data enough time to do so
How does data availability sampling work?
- We need a scalable way to ensure that when people want to download some data they can download it
- Data availability sampling involves selecting random parts of the data and trying to access them
- If you have access to most of the data, you think the data is available
- Reed-Solomon encoding enables you to reconstruct the entire dataset if 50% of the data is available
- This supports scaling as it doesn’t require 100% of the data to be available
- Attackers must reduce data availability to more than 50% to attack the network
- Every 10% of data downloaded reduces attacks by a factor of 10. This is a scalable way to ensure data availability
- Continue to expand the consensus layer
- Some data will be retained. After that, they will be trimmed
- Data will be available long enough to protect the network
- The amount of data available in proto-danksharding is less than the amount of data available in full danksharding
How much has the cost of storing data been reduced?
- Currently, Ethereum block size is between 50-100 kbs
- Can grow from 50 kbs to mb per block
- rollup will benefit from reduced costs
- In full danksharding, an order of magnitude increase
The difference between someone who pays a lot to perform computation and someone who pays a lot to store data
- Added additional fee parameters to create different marketplaces
- Vitalik wrote an article on Multidimensional EIP-1559
- When the data is sampled, there is no way to tell if the encoding is correct
- KZG promises are a way to hash polynomial functions and guarantee correct encoding
What is a trusted setup?
- Elliptic curve points that have a certain relationship must be set. This is one of the basic inputs of the KZG commitment scheme
- No one is allowed to know the actual relationship between them
- In a trusted setup, the setup is completely secure even if only one person did it correctly and everyone else colluded
When will EIP-4844 be deployed?
- Earlier this year, they wrote a proposal for this
- In the ETHDenver hackathon, they implemented it and have been testing it
- Two different branches: further development of client software and creation of testnets; development of trusted setups
- When done, they will include it as an EIP
- Proto-danksharding is a stepping stone to full implementation
- trying to get it done in the shanghai hard fork
What are the open issues in sharding if people want to contribute from a research/engineering perspective?
- Figure out the network for data availability sampling
- It would be nice to have more people with networking expertise
- Economic Challenges: How to Make a Good Proposer-Builder Separation Protocol? How to increase censorship resistance to bypass censorship builders?
- What would a better proof-of-stake design look like in the long run?
- Switching to Layer 2 and Adding Layer 2 Decentralization
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/vitalik-takes-you-to-understand-danksharding/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.