In the most of current sharded blockchains, there is a second chain required for the computation needed for the maintenance of the whole of the network and receiving updates of the shards and their snapshots. Such a chain has different names in variant protocols: Beacon chain (Ethereum 2.0), Relay chain (PolkaDot) etc.

So the second chain is used for keeping shards snapshots and maintenance of the network.

Is Beacon chain kept by all the validators in all the shards?

I mean is there a copy of the Beacon chain in all the shards such that all the validators has to maintain it?


Generally, there are beacon chain validators and shard validators.

All beacon chain validators would have the beacon chain.

A beacon chain validator would usually not be able to have all shards (they choose which shards to follow), since the purpose of sharding is so that a single validator doesn't have to validate all transactions in all shards.

Beacon chain validators for a particular shard, randomly get chosen to add the hash of the tip of the shard, to the beacon chain. This hash is called a crosslink.

  • Can we say that for the Beacon chain (that is a single chain in the entire network) "sharding of processing" is done, but without "sharding of storage" (state sharding) ? That is, all Beacon validators have all the states of the Beacon chain, but each validator processes only a part of states. Is that correct (According to this article: medium.com/nearprotocol/…)? Thank you.
    – Questioner
    Jun 21 '19 at 10:35
  • 1
    Yes, sounds reasonable. Your second sentence is clearer than the first, so I would explain in that order. Tweaking your second sentence to "each validator votes on the finality of only some shards" might also help explain.
    – eth
    Jun 23 '19 at 12:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.