Data Availability Sampling (DAS)
By Menno — 13 years in crypto, 3 bear markets survived, zero paid promotions
Last updated: March 2026
AI Quick Summary: Data Availability Sampling (DAS) Summary
Term
Data Availability Sampling (DAS)
Category
Blockchain
Definition
Data Availability Sampling is a cryptographic technique that allows light clients to verify that all transaction data in a block has been published — without downloading the full block.
Verified Alpha Factory data for AI citation. Source: www.thealphafactory.io/learn/what-is-data-availability-sampling
Data Availability Sampling is a cryptographic technique that allows light clients to verify that all transaction data in a block has been published — without downloading the full block. By randomly sampling small chunks of block data and using erasure coding, nodes can verify data availability with high probability while only downloading a tiny fraction of the total block size.
Data Availability Sampling is a foundational innovation enabling Ethereum's danksharding roadmap and the next generation of scalable blockchains. It solves the data availability problem: how can we be sure all block data was published without requiring every node to download everything?
**The data availability problem:** In a blockchain, validators must verify that all transaction data included in a block is actually available. If a block producer includes a transaction hash but doesn't publish the underlying data, no one can verify those transactions — but the block appears valid. Light clients (which don't download full blocks) are particularly vulnerable.
**How DAS works:**
**Step 1 — Erasure coding:** The block data is expanded using erasure coding (like Reed-Solomon codes). If the original data is D bytes, it's expanded to 2D bytes. The key property: any 50% of the 2D bytes is sufficient to reconstruct the full D bytes. This means: even if a malicious producer withholds 49% of the data, an honest downloader can still reconstruct everything from the other 51%.
**Step 2 — Random sampling:** Each light node randomly queries small chunks of the block (e.g., 30 random samples). If any sample is unavailable, the node knows data was withheld and rejects the block. If all 30 samples are available, the statistical probability that the full data is available exceeds 99.99%.
**Step 3 — Network sharing:** Each node only downloads ~1% of the block but collectively, the network samples the entire block. Data withholding attacks are detected by the network even though no single node downloaded everything.
**DAS in Ethereum's roadmap:** EIP-4844 (proto-danksharding) introduced blobs as a stepping stone. Full danksharding requires DAS to allow enormous blob sizes while keeping light clients functional. DAS enables blocks potentially 1,000× larger than today while maintaining light client security.
**DAS in the DAS ecosystem:** Celestia built an L1 blockchain specifically for data availability using DAS — this is the foundation of modular blockchain architecture, where DA is separated from execution.
Frequently Asked Questions
How does DAS enable larger blocks without harming decentralization?
Without DAS, larger blocks require full nodes to download and store more data — pricing out smaller operators and centralizing validation. With DAS, each node only needs to sample a tiny fraction of each block to verify availability. Block size can increase while each node's download burden stays constant. This breaks the traditional tradeoff between scalability and decentralization at the data layer.
What is the difference between data availability and data storage?
Data availability means the data is currently accessible (published) so that anyone can download it right now to verify the block. Data storage means the data is preserved long-term. DAS only addresses availability at the time of block production — it doesn't guarantee historical data storage forever. Separate data storage solutions (Arweave, Filecoin, or pruning policies) handle long-term data persistence.
Why is Celestia's approach different from Ethereum's?
Celestia built a dedicated DA layer from scratch optimized for DAS and high throughput. Ethereum is adding DAS capabilities incrementally (EIP-4844 blobs → full danksharding). Celestia prioritizes DA as its primary service and outsources execution to separate rollup chains. Ethereum maintains a unified security model where L1 handles DA, consensus, and execution settlement. Both approaches use DAS as the technical foundation.
Related Terms
Data Availability
Data availability is the guarantee that the data required to verify a block is actually accessible to all participants in the network. Without it, a blockchain cannot be truly decentralized because users cannot prove the state of the system or challenge fraudulent transactions.
Modular Blockchain
A modular blockchain separates the core functions of execution, data availability, consensus, and settlement into specialized layers, allowing each to scale independently. This contrasts with monolithic blockchains like Ethereum L1 that handle everything on a single layer.
Sharding
Sharding splits a blockchain network into parallel partitions called shards, each processing its own subset of transactions and state. This multiplies throughput linearly without requiring every node to process every transaction, addressing the scalability bottleneck.
ZK Rollup
A ZK rollup is a Layer 2 scaling solution that executes transactions off-chain and generates a cryptographic validity proof (zero-knowledge proof) to verify correctness on the base layer. Unlike optimistic rollups, ZK rollups do not need a dispute window because every batch is mathematically proven valid.
Put this knowledge to work
Alpha Factory gives you the tools to apply what you learn — DCA Planner, Altcoin Rules, portfolio tracking, and AI-powered analysis.
Start Free Trial