Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve succinct operator to orchestrate proof requests #13

Open
dndll opened this issue Feb 20, 2024 · 0 comments
Open

Improve succinct operator to orchestrate proof requests #13

dndll opened this issue Feb 20, 2024 · 0 comments

Comments

@dndll
Copy link
Member

dndll commented Feb 20, 2024

Description

At the moment, the succinct operator follows a one-shot style command system. This experience is not seamless or efficient when it comes to make requests for proofs. To integrate well with DA, we need to introduce a system where we can automatically request proofs once certain criteria is met.

Why?

Consider the DA use case, where arbitrary users submit transactions, and those transactions need to be proven on ethereum. This requires the following steps:

  • submit transaction
  • request to prove
  • prove
  • relay to verifier
  • verify
  • emit event

If we do this for each blob submission, it's likely this is expensive. At the moment the verification circuit batches transactions/receipts to be proven in batches of 128. We can implement a queue mechanism which will drain the queue for proving on a few triggers:

  • when the queue is full
  • a configurable timeslot
  • hard override

For 2 and 3, we have to expand the queue with empty transactions, wasting some relay space. We do gain proving time, since the default proofs are ignored.

There is also an additional dimension where the light client head would need to be synced to support the latest proof. If we can somehow amortize this into one verification then that would also be ideal, until now, the queue should trigger a sync for the latest head, then drain the queue for proving, unless the queue items could be proven for the current head.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant