Skip to content

Cysic Auction Mechanism

1. Motivation

Efficient task assignment in decentralized computation requires a fair, incentive-compatible, and penalty-enforced bidding system. The auction mechanism ensures that tasks are allocated to provers with sufficient computational resources, while maintaining economic alignment between requesters, provers, and verifiers.

2. Design Goals

  • Fairness: Guarantee that tasks are allocated based on competitive bidding under a capped maximum price.
  • Efficiency: Encourage timely completion by rewarding faster provers with higher payouts.
  • Security: Ensure that dishonest or underperforming provers are penalized through slashing.
  • Incentive Alignment: Incorporate token reserves into the reward distribution, aligning long-term commitment with higher earnings.

3. Task Definition

A task is published by the Requester with the following parameters:

  • Task Difficulty (task_difficulty): Expressed in cycles.
  • Maximum Acceptable Bid (bid_max): The upper bound of unit price, in CYS per million cycles.
  • Task Deadline (task_ddl): The maximum allowable computation time.

4. Prover Requirements

  • Token Reservation: Each prover must lock a minimum amount of tokens to be eligible.
  • Bidding Configuration: Provers specify their bidding price in the configuration file, which serves as the reference for subsequent auctions.
  • Performance Evaluation: By running the client software, a prover’s computational capacity is automatically benchmarked.

5. Auction Flow

  1. Task Broadcast: The requester publishes the task, which is propagated to all registered provers.
  2. Capability Assessment: Each prover determines whether it can complete the task within task_ddl.
  3. Bid Submission: Provers submit bids reflecting their desired compensation.
    • Bids exceeding bid_max are automatically discarded.
  4. Bid Selection: Upon bidding closure, the system sorts valid bids by price.
    • The two winning provers are chosen starting from the second-lowest bid.
    • If two provers submit the same price, the more reserved one will be chosen.
    • The selected bid price (bid_select) is the lowest bid among these winners.

6. Reward Distribution

6.1 Prover Rewards

The total reward pool for provers is defined as:

\[ task\_reward\_prover = bid\_select \times task\_difficulty \times 80\% \]

This reward is distributed among the three winning provers based on the verification results from verifiers.

6.2 Failure and Slashing

  • If a prover fails to meet the deadline, the task is reassigned to a backup prover.

  • The failing prover incurs a penalty:

    \[ slash = \beta \times bid\_max \times task\_difficulty \]

    where \(\beta = 1\).

6.3 Verifier Rewards

  • 20 verifiers are randomly selected to validate the proof. The selection is determined by the block hash at task creation time, ensuring unpredictable and tamper-resistant randomness.

  • The verification task concludes once 60% of the selected verifiers (i.e., at least 12) have submitted their results. Only those who successfully submit before the task closes will share the reward:

    \[ verifier\_reward = \frac{bid\_select \times task\_difficulty \times 20\%}{n_{submitted}} \]

    where \(n_{submitted}\) is the number of verifiers who submitted in time.

  • Note: Since rewards are distributed only among verifiers who submit before task closure, maintaining low-latency network connectivity is essential for maximizing earnings.

7. Reserve-Weighted Incentives

To strengthen long-term commitment and discourage opportunistic behavior, prover rewards are further adjusted by token reserves:

\[ final\_task\_reward\_prover = task\_reward\_per\_prover \times \big(1 + \gamma \times R\big) \]

Where:

  • \(\gamma\): Initially set to 0.25, but subject to adjustment in the future depending on the overall reserve distribution.
  • \(R = \frac{reserve\_token}{total\_reserve\_token}\): relative reserve ratio.

This ensures that provers with larger token commitments receive proportionally higher rewards.

8. Security Considerations

  • Sybil Resistance: Token reservation reduces the risk of Sybil attacks by requiring economic commitment.
  • Collusion Prevention: Pricing determined by the second-lowest bid mitigates collusion among provers.
  • Reliability Enforcement: Slashing enforces accountability, ensuring underperforming provers bear economic costs.
  • Verifier Integrity: Distributing rewards among multiple verifiers enhances the robustness of proof validation.