R&D: Multi-prover Implementation

[orkunkilic]

We need to explore and develop a research prototype for implementing a multi-prover system. The goal is to make the proof generation process faster and resistant to the proof system bugs by distributing it across multiple independent provers.

Objectives:

• Research Existing Approaches: Investigate current multi-prover schemes, their pros and cons, and how they can be adapted or optimized for our system.
  • Scroll's SGX Prover
• Define Roles & Responsibilities: Outline how each prover would contribute to the generation and aggregation of zk proofs such as light client proof generation and BitVM proofs.
• Synchronization & Consensus: Ensure the provers stay in sync, avoid data inconsistencies, and maintain consensus on final zk proofs.
• Performance Metrics: Define success criteria, including benchmarks on proof generation times, network overhead, and resource utilization.
• Research Data Availability: Find out the optimal approach for posting state diffs on Bitcoin.

Expected Outcomes:

• A proof-of-concept implementation of a multi-prover Citrea batch proofs.
• A report on the performance trade-offs and feasibility of the approach.
• Security considerations and potential risk mitigations.

Dependencies:

• Current batch proof and light client proof implementations in Rust.
• Coordination mechanism for provers

Priority: Mid