Data Availability Committee (DAC)

Protocols using a DAC often fall under the Validium category of scaling solutions. Key differentiators include:

• Security Model: Relies on a committee's honesty for data availability, not cryptographic proofs on-chain.
• Cost & Throughput: Offers the lowest fees and highest throughput as no data is posted to L1.
• Trade-off: Introduces a data availability risk; if the committee censors or fails, funds cannot be withdrawn. This contrasts with Rollups which guarantee data is on Ethereum.

A modern trend is combining DACs with EigenDA, a restaked data availability layer built on EigenLayer. In this model:

• The DAC provides fast attestations and initial liveness guarantees.
• EigenDA acts as a secondary, cryptoeconomically secured layer where data is eventually posted and secured by restakers. This hybrid approach aims to blend the speed and cost of a committee with the robust, decentralized security of a proof-of-stake network.

A DAC operates on an explicit trust model, where users must trust the honesty and liveness of committee members. This contrasts with cryptographic data availability solutions like Data Availability Sampling (DAS) or erasure coding with validity proofs, which provide trust-minimized, mathematical guarantees that data is available. The security of a DAC-based system is only as strong as the collective integrity of its members.

The security of a DAC hinges on its members' reputation, economic stake, and legal identity. Key considerations include:

• Member Selection: Are members well-known, reputable institutions or anonymous entities?
• Collusion Resistance: What prevents a majority of members from colluding to withhold data?
• Slashing & Bonds: Is there a cryptographic slashing mechanism for misbehavior, or are penalties based on legal contracts and reputational damage?

The primary failure mode is data withholding, where the committee refuses to provide the data needed to reconstruct the chain state. This can lead to:

• Funds being frozen if users cannot generate fraud proofs or withdraw assets.
• Effective censorship of transactions.
• The risk is unquantifiable; there is no cryptographic way to prove data is being withheld until an attempt to retrieve it fails.

Users must assume the DAC maintains high liveness (always online and responsive). If a critical mass of members goes offline, data becomes unavailable, halting chain operations or withdrawals. This contrasts with peer-to-peer networks where data redundancy across thousands of nodes provides robust liveness. DACs often implement service level agreements (SLAs) rather than protocol-enforced liveness guarantees.

A Validium is a specific scaling architecture that uses a DAC for data availability while executing transactions off-chain and using zero-knowledge proofs for validity. Its security model is a hybrid:

• Validity is guaranteed cryptographically by ZK proofs.
• Availability is trusted to the DAC.
• This trade-off enables higher throughput and lower costs than a ZK Rollup, but inherits all the trust assumptions associated with the committee.

Many projects begin with a DAC as a pragmatic launch solution and outline a path to decentralize data availability over time. This transition may involve:

• Increasing the number and diversity of committee members.
• Introducing staking and slashing mechanisms.
• Gradually migrating to a hybrid model that incorporates elements of cryptographic DA sampling alongside committee signatures.
• The end goal is often full reliance on a decentralized DA layer like Celestia, EigenDA, or Ethereum's danksharding.

A Layer 2 scaling solution where transaction execution occurs off-chain and proofs are posted on-chain, but data availability is handled by a committee or DAC instead of the base layer (L1). This trades some decentralization for significantly lower transaction costs.

• Use case: High-throughput applications where cost is critical and a trusted DAC is acceptable.
• Contrast with Rollups: Rollups post data to L1; Validiums do not, relying on the DAC's attestations.

A specialized blockchain layer whose primary function is to order and guarantee the availability of transaction data for other execution layers. It decouples data availability from execution.

• Examples: Celestia is a sovereign data availability layer. Ethereum, post-EIP-4844 (proto-danksharding), incorporates data availability features.
• Purpose: Allows rollups and other L2s to post data cheaply and securely, knowing it is permanently accessible for verification.

A key data redundancy technique used in data availability schemes. It expands the original data with parity chunks, so the full data can be reconstructed even if some chunks are missing or withheld.

• Role in DAS: Makes data availability sampling feasible. A node only needs to sample a few chunks to be confident the entire data is available.
• Fault tolerance: Protects against data withholding attacks by malicious block producers.

A cryptographic economic security mechanism where members of a Data Availability Committee have stake (e.g., tokens) that can be destroyed (slashed) if they sign an availability attestation for data that later proves to be unavailable.

• Security model: Aligns committee incentives with honesty. Dishonest behavior has a direct financial cost.
• Limitation: Requires a robust fraud proof system to detect and prove data was withheld.

A user-configurable architecture, pioneered by StarkEx, that lets users choose for each transaction whether data availability is secured on-chain (like a ZK-Rollup) or off-chain by a DAC (like a Validium).

• Flexibility: Users can opt for maximum security (L1) for high-value assets or lower cost (DAC) for high-frequency trades.
• Example: Immutable X and Sorare utilize this model, providing users with a choice.