Skip to main content

Consensus Mechanism

Quilibrium implements a novel consensus mechanism called Proof of Meaningful Work that ensures network security while requiring participants to contribute valuable computational resources.

Overview​

Unlike traditional Proof of Work systems that perform arbitrary hash calculations, Quilibrium's consensus requires provers to execute meaningful computations that benefit the network and its users.

Proof of Meaningful Work​

Core Principles​

  1. Computational Contribution: Provers must execute real computations requested by network users
  2. Storage Contribution: Provers must provide succinctly verifiable proof of storage
  3. Verifiable Results: All computational outputs must be cryptographically verifiable
  4. Economic Incentives: Provers are rewarded for contributing computational resources
  5. Security Through Work: The computational effort required provides network security

Prover Network​

The consensus mechanism relies on a distributed network of provers:

Prover Roles​

  • Computation Execution: Running QCL (Quilibrium Compute Language) programs
  • Data Validation: Verifying hypergraph operations and state transitions
  • Network Maintenance: Participating in frame consensus and production
  • Storage Provision: Maintaining distributed state across shards

Prover Selection​

Provers are selected for specific tasks based on:

  • Shard Assignment: The shards the provers have allocated for themselves
  • Computational Capacity: Available capabilities and storage

Frame-Based Architecture​

Frame Structure​

The protocol operates in discrete time periods called frames:

  • Duration: Fixed time intervals (ten seconds)
  • Ordering: Sequential numbering (on conflicts) ensures deterministic state progression
  • Finality: Frames become immutable once consensus is reached

Frame Consensus Process​

  1. Transaction Collection: Provers gather pending transactions from the mixnet
  2. Computation Execution: Provers execute computations and prepare results
  3. Proof Generation: Cryptographic proofs are created for all operations
  4. Consensus Round: Provers participate in Byzantine Fault Tolerant consensus
  5. Frame Commitment: Agreed-upon frame is committed to the global state

State Transitions​

Each frame represents a state transition of the network:

State(n) + Transactions(frame_n) = State(n+1)

Consensus Algorithm Details​

This section is omitted until the release of 2.1

Economic Model​

Reward Distribution​

Provers receive rewards for:

  • Computation Execution: Payment for running user computations
  • Frame Production: Rewards for successfully proposing frames

Penalty System​

Malicious or unreliable behavior is penalized through:

  • Slashing: Seniority decays on missed proof intervals
  • Reputation: Seniority dictates conflict resolution for shard joins
  • Ejection: Removal from prover set for serious violations, complete loss of seniority

Security Guarantees​

Attack Resistance​

The consensus mechanism provides resistance against:

  • 51% Attacks: Proof system ensures one honest prover is sufficient to remove malicious majorities
  • Nothing at Stake: Economic penalties (through loss of seniority and eviction) for signing conflicting frames
  • Long Range Attacks: Single-slot finality under normal conditions, checkpoint finality under probabilistic conditions prevents historical rewrites
  • Eclipse Attacks: Diverse peer connections and reputation systems

Finality​

Frames achieve different levels of finality:

  • Soft Finality: Immediate probabilistic finality after consensus
  • Hard Finality: Immediate finality under normal conditions, probabilistic finality after checkpoint confirmation
  • Economic Finality: Cost of reversal exceeds potential gain

Performance Characteristics​

Throughput​

  • Transaction Processing: Thousands of transactions per shard frame
  • Computation Capacity: Scales with prover network size
  • Data Storage: Distributed across multiple shards

Latency​

  • Frame Time: Predictable intervals for state updates
  • Confirmation Time: Fast confirmations via mixnet
  • Finality Time: Deterministic finality within known bounds

Implementation Details​

Prover Coordination​

Provers coordinate through:

  • Gossip Protocol: Efficient propagation of transactions and proofs
  • Leader Election: Verifiably random leadership selection for frame proposal
  • Synchronization: Clock synchronization for frame timing

State Management​

  • Sharding: Horizontal partitioning of state for scalability
  • Replication: Multiple copies of critical state for availability
  • Consistency: Strong consistency within shards, eventual consistency across shards