Skip to main content

Theory 2: Sharding, data availability sampling, and cryptography primitives

Readings:#

  • Why Sharding is Great: demystifying the technical properties Sharding is a key component of the future of Ethereum Layer 1 and is one of the most misunderstood concepts in crypto (largely due to analogizing with database sharding). This post is a handy guide on what exactly sharding gives us and why it's crucial to add to Layer 1.
  • Kate commitments (warning: Math!) Commitments to an ordered set of data that are cheap to prove inclusion on are a very key primitive in Layer 1 blockchains. Currently, Merkle Patricia Trees are used almost everywhere. Research into a more efficient data structure has led us to Kate commitments, which have some really nice properties and are likely to be a foundational cryptographic data structure in Layer 1 in the future.
  • Data Availability Sampling with Kate commitments Data availability sampling is the other half (other than sharding) of the known large Layer 1 changes to happen in the near future. It's a crucial scaling technology as well!
  • (optional) Dankrad and Vitalik on Verkle Tries Verkle tries are a proposed data structure that are likely to end up in Eth1x to reduce witness size for stateless clients (don't worry if this makes no sense to you now). They're a great example in the short-term of how Kate commitments might be used.

Intuitions to build:#

  • why build sharding into core consensus? how are shards different from rollups?
  • what is data availability sampling? why is data availability important?
  • what are Kate commitments good for?