DeepSummary
The podcast episode discusses data availability sampling (DAS) and danksharding, which are concepts relevant to blockchain scaling, particularly for Ethereum. DAS is a low-overhead technique that uses random sampling of data to ensure that all necessary blockchain data has been made available to nodes without straining the network. Danksharding is a series of planned Ethereum upgrades that incorporates DAS and aims to achieve full scaling for Ethereum.
The guests, Valeria Nikolaenko and Dan Boneh, explain how DAS works and its role in danksharding. They discuss the current proposal for danksharding and their proposed improvements, which involve using bivariate polynomial interpolation instead of univariate interpolation to reduce the amount of data needed for reconstruction from 75% to 25%.
The episode covers various technical details related to danksharding, such as erasure coding, polynomial commitments, and the trusted setup ceremony. It also touches on the timeline for the Ethereum upgrades, with protodanksharding (EIP-4844) planned for later this year as a stepping stone towards full danksharding.
Key Episodes Takeaways
- Data availability sampling (DAS) is a low-overhead technique that uses random sampling of data to ensure blockchain data availability without straining the network.
- Danksharding is a series of planned Ethereum upgrades that incorporates DAS and aims to achieve full scaling for Ethereum.
- The current proposal for danksharding uses univariate polynomial interpolation, but the guests propose using bivariate polynomial interpolation to reduce the amount of data needed for reconstruction from 75% to 25%.
- Erasure coding and polynomial commitments are key concepts in danksharding, allowing data to be split and distributed across validators while enabling reconstruction.
- Protodanksharding (EIP-4844) is a stepping stone towards full danksharding and is planned for deployment on Ethereum later this year.
- Danksharding and the need for efficient bivariate polynomial interpolation are driving new research areas in cryptography and computer science.
- The ultimate goal of danksharding is to enable a scalable version of Ethereum where rollups and other layer 2 solutions are much cheaper to use.
- The episode covers various technical details and mathematical concepts related to danksharding and DAS, such as erasure coding, polynomial commitments, and trusted setup ceremonies.
Top Episodes Quotes
- “Erasure coding is used everywhere, like, communication networks wouldn't be able to work without erasure coding.“ by Dankrad Feist
- “So to me, this is really one beautiful. In that the blockchain Ethereum dank sharding is creating this new area of research, or at least prioritizing this new area of research, showing, you know, we really need better algorithms, new algorithms, you know, efficient algorithms to do bivariate polynomial interpolation.“ by Dankrad Feist
- “If you reconstruction only requests 95% of the data, you do less samples and your assurance, the false positive rate goes down quicker than if you only have a reconstruction algorithm that requires 75% of the data.“ by Dan Boneh
- “The upshot is a scalable version of Ethereum where rollups are much cheaper to use than they are today.“ by Dankrad Feist
Entities
Company
Person
Organization
Concept
Product
Book
Episode Information
web3 with a16z crypto
a16z crypto, Sonal Chokshi, Chris Dixon
5/5/23
with @lera_banda @danboneh @rhhackett
This episode introduces concepts behind -- and applications of -- data availability sampling (DAS), a key piece of the puzzle that could help blockchains like Ethereum achieve full scaling. A low-overhead technique that uses random sampling of data to ensure that all necessary blockchain data has been made available to nodes without straining the network, DAS features heavily in a series of planned Ethereum upgrades called "Danksharding" [named to reference Ethereum Foundation researcher Dankrad Feist]. The next Danksharding milestone for Ethereum is an upgrade called EIP-4844, known as "Protodanksharding" [also named to reference Ethereum researcher Protolambda, now at OP Labs] -- which is planned for later this year.
Our expert guests include:
- Valeria Nikolaenko, a16z crypto research partner
- Dan Boneh, Stanford cryptography professor and a16z crypto senior research advisor
...who discuss their recent piece on data availability sampling and Danksharding -- including a proposal they put forward to the improve current plans for upgrading Ethereum -- with Robert Hackett (a16z crypto features editor and head of special projects), based on a live conversation that took place recently on Twitter Spaces.
links to pieces/ topics referenced in this episode:
- on data availability sampling and danksharding, an overview and proposal / Valeria Nikolaenko & Dan Boneh
- on Proto-Danksharding, a technical analysis / grizzly-answer-991
- EIP-4844: the specification of ProtoDankSharding / Ethereum Foundation
- Proto-Danksharding FAQ / Vitalik Buterin
- Danksharding workshop video / Devcon (Oct 2022)
- 2D data availability with Kate commitments discussion on commitments' expansion / Ethereum Research Forum
- a note on data availability and erasure coding / Github
---
None of this is investment, business, legal, or tax advice; please see a16z.com/disclosures for more important information – including a link to a list of our investments.