8

I was questioning this for a while. Having parachains are generally good for security and the ecosystem.

As seen in the Polkadot whitepaper, 2nd order Relay-chain is mentioned.

Polkadot whitepaper

I believe it is not possible since I haven't seen any example of it or a chain that does this.

So, I have a few questions about this implementation.

  • What are the drawbacks?
  • Parachains don't have block finalization, it's being done in the relay chain. Would adding a 2nd order Relay-chain with block finalization be possible at all?
Improve this question
2
  • Parachains do have block finalisation - when the relay block is finalised then the associated parachain block is finalised. There's no problems with recursive relay chains from that perspective.
    – Squirrel
    Commented Apr 10, 2022 at 6:52
  • 2
    It could be done, but as you point out, the approval-checking and disputes mechanisms would have to be done differently as the 2nd order relay chain doesn't have authority over its own finalization.
    – rob
    Commented Apr 10, 2022 at 20:07

2 Answers 2

Reset to default
5

What are the drawbacks?

The most notable drawback to continually nesting relay chains is the increased delay of things like finalization, cross-chain messages, and other things from the outer chains to the main Polkadot relay chain.

For example, imagine the block time of all chains is 6 seconds. If you have 3 nested relay chains, you would expect that any cross-chain message from a parachain on the 3rd layer to the main Polkadot relay chain would take 18 seconds or more, since each layer would need to produce a block to the lower layer with the message in it.

Luckily, parachains which are in active communication can position themselves close together to reduce this delay, but as mentioned, there would still be lag between block production on the outer edges of the network, and block finalization at the main Polkadot relay chain.

I believe it is not possible since I haven't seen any example of it or a chain that does this.

As Rob mentions, the current implementation of the parachains protocol does not have all the required logic to supported nested relay chains, which is why you don't see them.

But you should expect that in the future, this will be made possible.

Improve this answer
1
  • What factors might be hindering the development of the nested relay chain? @shawn-tabrizi
    – Kresna
    Commented Feb 23, 2023 at 12:39
0

It seems to me that the drawback is that one would need a token to stake to secure the set of para-para-chains. The safety of all the para-para-chains would depend upon DOT and this additional PoS token. I.e. there might be a limit to how much economy you could have underneath a relay-relay chain before it becomes viable for attacks on it (it really would depend on the PoS token price).

An alternative scaling solution might be to validate ZK proofs on the para-chain. I think Manta is an example of this route.

I'm sure these are not the only two options in the design space! I'm sure we will see a lot of innovation trialing out the various approaches.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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