Data Availability Committees (DACs) excel at providing ultra-low-cost data availability by using a small, permissioned set of trusted entities. This approach, used by networks like Celestia's Sovereign Rollups and Arbitrum Nova, can reduce transaction costs by over 90% compared to posting all data directly to Ethereum L1. The trade-off is a weaker security assumption, as users must trust the committee's liveness and honesty.
LABS
Comparisons
Data Availability Committees (DACs) vs Pure On-chain Data Availability
A technical analysis comparing the trusted, permissioned model of Data Availability Committees with the cryptographically secured, trust-minimized model of pure on-chain data availability for rollup architectures.
Chainscore © 2026
introduction
THE ANALYSIS
Introduction: The Core Trade-off in Rollup Data Availability
The choice between Data Availability Committees and pure on-chain storage defines your rollup's security model, cost, and scalability.
Pure On-chain DA takes a different approach by publishing all transaction data directly to a base layer like Ethereum, Bitcoin, or Solana. This results in the highest security and censorship resistance, as seen with Arbitrum One and OP Mainnet, but at a significantly higher cost. This cost scales with the base layer's gas fees, which can be volatile and prohibitive for high-throughput applications.
The key trade-off: If your priority is minimizing cost for a high-TPS consumer app and you can accept a trust assumption, a DAC-powered solution like Mantle Network is compelling. If you prioritize maximizing security and decentralization for a high-value DeFi protocol, choose a rollup with pure on-chain DA to Ethereum or another robust L1.
tldr-summary
Data Availability Committees (DACs) vs Pure On-chain DA
TL;DR: Key Differentiators at a Glance
A high-level comparison of the two dominant data availability models, highlighting their core trade-offs in security, cost, and decentralization.
01
DACs: Lower Cost & Higher Throughput
Specific advantage: Offloads data to a small, permissioned committee, drastically reducing on-chain footprint. This matters for high-frequency applications like gaming or social feeds where cost-per-transaction is critical. For example, Celestia's rollups using DACs can achieve sub-cent fees compared to Ethereum's $1+ L2 fees.
02
DACs: Practical for Early-Stage Scaling
Specific advantage: Enables rapid deployment and iteration for new L2s and appchains without the full security overhead. This matters for prototyping and MVPs where time-to-market and capital efficiency are paramount. Projects like Mantle Network initially leveraged a DAC to bootstrap their ecosystem before a planned migration to a more decentralized DA layer.
03
Pure On-chain DA: Unmatched Security & Censorship Resistance
Specific advantage: Data is published directly to a base layer (e.g., Ethereum, Bitcoin) with thousands of validating nodes. This matters for high-value DeFi and institutional assets where the security of billions in TVL is non-negotiable. The data availability sampling (DAS) in pure models like Ethereum danksharding provides cryptographic guarantees without trusted parties.
04
Pure On-chain DA: Strongest Decentralization & Composability
Specific advantage: Inherits the full trust assumptions and network effects of the underlying L1. This matters for interoperability and maximal security; any Ethereum L2 using Ethereum for DA (e.g., Arbitrum, Optimism, zkSync) can be trustlessly verified by anyone. It creates a unified, sovereign-grade security pool for the entire ecosystem.
DATA AVAILABILITY ARCHITECTURE
Head-to-Head Feature Comparison: DACs vs On-chain DA
Direct comparison of security, cost, and performance for rollup data availability.
| Metric | Data Availability Committee (DAC) | Pure On-chain DA (e.g., Ethereum) |
|---|---|---|
Security Model | Multi-signature Trust | Cryptographic & Economic |
Data Guarantee | ||
Cost per 100 KB Blob | $0.01 - $0.10 | $1.00 - $5.00 |
Latency to Post Data | < 1 sec | ~12 sec (Ethereum block time) |
Censorship Resistance | Committee-dependent | Permissionless |
Active Implementations | Celestia, Avail, EigenDA | Ethereum, Polygon Avail |
pros-cons-a
DACs vs. Pure On-chain DA
Pros and Cons: Data Availability Committees (DACs)
Key strengths and trade-offs at a glance for CTOs and architects choosing a data availability layer.
01
DACs: Lower Cost & Higher Throughput
Specific advantage: Significantly lower transaction fees and higher TPS by avoiding full on-chain data posting. This matters for high-frequency applications like gaming or micro-transactions where cost and speed are primary constraints. Example: A rollup using a DAC like Celestia's Blobstream or EigenDA can achieve costs under $0.001 per transaction.
02
DACs: Faster Finality for Rollups
Specific advantage: Near-instant data attestation from a known, permissioned committee. This matters for optimistic rollups needing to minimize their dispute window (e.g., from 7 days to hours) or for any application requiring rapid proof confirmation. Protocols like Arbitrum AnyTrust rely on this for faster withdrawal times.
03
DACs: Trust & Security Assumption
Specific weakness: Relies on the honesty of a permissioned set of nodes (e.g., 10-100 known entities). This matters for maximally decentralized applications where even 1-of-N collusion risk is unacceptable. A breach compromises chain validity, unlike pure on-chain DA which inherits L1 security.
04
DACs: Limited Ecosystem Interoperability
Specific weakness: Data is not natively verifiable by all L1 validators, fragmenting liquidity and composability. This matters for DeFi protocols requiring universal asset proofs across chains. A bridge or light client must be trusted to relay DAC attestations, adding complexity vs. native Ethereum blobs.
05
Pure On-chain DA: Maximum Security
Specific advantage: Data availability is guaranteed by the full validator set of the base layer (e.g., Ethereum's ~1M validators). This matters for high-value, institutional DeFi where security is non-negotiable. Protocols like StarkNet and zkSync Era use Ethereum calldata/blobs for canonical settlement.
06
Pure On-chain DA: Universal Composability
Specific advantage: Data is published in a globally accessible, canonical location. This matters for cross-rollup interoperability and shared liquidity pools. Any smart contract or prover can directly verify the data without extra trust assumptions, enabling seamless integration across the L2 ecosystem.
pros-cons-b
DACs vs. On-Chain DA
Pros and Cons: Pure On-chain Data Availability
Key strengths and trade-offs at a glance for Data Availability Committees (DACs) and Pure On-chain Data Availability (DA).
01
DACs: Lower Cost & Higher Throughput
Specific advantage: Offloading data storage to a small, permissioned committee drastically reduces gas fees and increases throughput. This matters for high-frequency applications like gaming or perp DEXs where sub-cent transaction costs are critical. For example, a DAC can enable 10,000+ TPS at a fraction of the cost of posting the same data to Ethereum.
02
DACs: Centralization & Trust Assumptions
Specific disadvantage: Relies on a small, known set of entities (e.g., 10-20 members) for data availability. This introduces trust assumptions and censorship risk, as the committee can collude or fail. This matters for decentralized finance (DeFi) protocols where the security of billions in TVL cannot depend on a handful of actors. It's a trade-off of scalability for decentralization.
03
Pure On-chain DA: Unmatched Security & Censorship Resistance
Specific advantage: Data is posted directly to a base layer like Ethereum, inheriting its full consensus-level security and decentralization. This matters for sovereign chains, high-value bridges, and institutional DeFi where data liveness must be guaranteed by thousands of globally distributed validators, not a committee. The security budget is the underlying L1's (e.g., Ethereum's ~$50B+ staked ETH).
04
Pure On-chain DA: High Cost & Scalability Limits
Specific disadvantage: Paying for full calldata storage on a secure L1 is expensive and throughput-constrained. This matters for mass-market dApps and rollups needing low fees. For instance, posting 1 MB of data to Ethereum can cost over $1,000 during congestion, directly translating to high rollup fees. This creates a hard scalability ceiling tied to base layer capacity.
CHOOSE YOUR PRIORITY
When to Choose: Decision Framework by Use Case
Pure On-chain DA for High-Throughput Apps
Verdict: The default choice for maximum scale. Strengths: Offers the highest theoretical throughput, as seen on Solana (50k+ TPS) and Sui. Data is immediately available to all validators, enabling parallel execution and sub-second finality. This is critical for order-book DEXs like Phoenix, high-frequency trading, and social apps with massive micro-transactions. Trade-offs: Demands extremely high node hardware specs, leading to centralization pressures. Full nodes become expensive to run.
DACs for High-Throughput Apps
Verdict: A pragmatic scaling bridge for EVM ecosystems. Strengths: Enables high TPS (e.g., 10k+ on Arbitrum Nova) while keeping transaction fees extremely low ($0.001-$0.01). By moving data availability off the base layer to a trusted committee (e.g., using EigenDA, Celestia, or a custom DAC), it drastically reduces L1 calldata costs. Ideal for scaling gaming, high-volume NFT minting, and micro-payments on rollups. Trade-offs: Introduces a trust assumption in the committee members (typically 7-10 reputable entities). Not suitable for ultra-high-value DeFi where cryptoeconomic security is non-negotiable.
verdict
THE ANALYSIS
Verdict and Final Recommendation
A final assessment of Data Availability Committees versus pure on-chain DA, framed by the fundamental trade-off between scalability and trust.
Data Availability Committees (DACs) excel at providing high-throughput, low-cost data availability by leveraging a small, permissioned set of trusted entities. This off-chain model drastically reduces the consensus overhead required by pure on-chain systems, enabling transaction costs under $0.01 and throughput exceeding 10,000 TPS, as seen in implementations like Celestia's Sovereign Rollups or Polygon Avail's early testnets. The primary benefit is scalability without the immediate need for massive validator decentralization.
Pure On-chain Data Availability takes a fundamentally different approach by enforcing data availability through cryptographic proofs and a large, decentralized validator set, as pioneered by Ethereum's danksharding (Proto-Danksharding/EIP-4844). This results in a stronger security model—inheriting the base layer's trust assumptions—but introduces a significant trade-off in cost and capacity. While Ethereum's full danksharding targets ~1.3 MB/s, current blob-carrying transactions are still more expensive than DAC-settled transactions.
The key architectural divergence is trust minimization versus performance optimization. DACs operate on an optimistic security model, assuming committee members are honest, with fraud proofs or slashing as a backstop. Pure on-chain DA provides cryptographically guaranteed availability, making it the gold standard for high-value, security-critical applications like L1 bridges or restaking protocols where trust cannot be outsourced.
Consider the ecosystem and tooling maturity. Building with a DAC often means integrating with a specific stack (e.g., EigenDA, Celestia, Avail) which offers tailored SDKs and faster iteration. Choosing pure on-chain DA, typically via Ethereum, grants immediate access to the largest developer ecosystem, battle-tested clients like Geth and Nethermind, and seamless composability with existing DeFi TVL, which exceeds $50 billion.
The final decision is use-case driven. Choose a DAC if your priority is ultra-low transaction fees, maximum scalability for a consumer app, or you are building a sovereign rollup chain that values execution flexibility. Opt for pure on-chain DA if your priority is maximizing security and decentralization for high-value assets, requiring the strongest possible audit trail, or building a protocol that must integrate directly with the Ethereum ecosystem's liquidity and users.
ENQUIRY
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.
NDA Protected
Confidentiality24h Response
Time to ValueDirectly to Engineering Team
No Sales Fluff10+
Protocols Shipped
$20M+
TVL Overall