Ethereum's core product is finality. It provides a globally trusted, immutable ledger for state commitments. Execution is a commodity that scales horizontally on L2s like Arbitrum and Optimism.
the-ethereum-roadmap-merge-surge-verge
Blog
Why Ethereum Treats Data as a First-Class Resource
An analysis of how Ethereum's post-Merge roadmap, specifically The Surge, redefines scaling by making data availability a native, monetizable layer-1 primitive, challenging external DA layers like Celestia.
introduction
THE DATA
The Contrarian Take: Ethereum is a Data Availability Layer
Ethereum's primary value is securing high-value state commitments, not executing transactions.
Data availability is the bottleneck. The cost of posting transaction data to Ethereum, via blobs, is the primary constraint for L2 scaling. This makes calldata pricing the key economic lever for the entire rollup ecosystem.
EIP-4844 (Proto-Danksharding) proves the thesis. The upgrade introduced a dedicated blob space market, decoupling L2 data costs from mainnet execution gas. This cemented Ethereum's role as a settlement and data layer.
Evidence: Post-EIP-4844, L2 transaction fees dropped 90%. The blob gas fee market now directly dictates the economic throughput of chains like Base and zkSync, validating the data-first resource model.
thesis-statement
THE SCARCITY SHIFT
Core Thesis: Data is the New Block Space
Ethereum's primary constraint has shifted from raw computation to the cost and availability of on-chain data.
Execution is cheap, data is expensive. Rollups like Arbitrum and Optimism handle computation off-chain, but their security depends on posting data to Ethereum's call data. This makes data availability the new bottleneck for scaling.
EIP-4844 (Proto-Danksharding) formalizes this. It creates a dedicated blob space market separate from execution gas, allowing L2s to post data at ~90% lower cost without congesting the main chain.
The market values data permanence. Permanent storage on-chain (via calldata) costs 20-40x more than temporary blob storage, proving that applications pay a premium for guaranteed, long-term data availability.
Evidence: Post-EIP-4844, Base and zkSync now post >90% of transactions as blobs, reducing L2 fees by an order of magnitude and validating the data-centric scaling thesis.
key-trends
WHY DATA IS THE NEW OIL
The Data-Centric Roadmap: Three Strategic Pivots
Ethereum's evolution from a monolithic compute engine to a modular settlement layer is a fundamental bet on data as the primary resource for security and scalability.
01
The Problem: Monolithic Chains Hit a Data Wall
Packing execution, settlement, consensus, and data availability on one chain creates a fundamental bottleneck. Every node must process every transaction, leading to ~15-45 TPS and $10+ gas fees during congestion. Scaling compute alone (sharding execution) fails without scaling data.
~15-45
Peak TPS
$10+
Congestion Fee
02
The Solution: Proto-Danksharding & Data Availability Sampling
EIP-4844 introduces blob-carrying transactions, creating a dedicated, cheap data market separate from execution gas. Validators use Data Availability Sampling (DAS) to cryptographically confirm data is available without downloading it all, enabling secure scaling for L2s like Arbitrum and Optimism.\n- Key Benefit 1: ~100x cheaper data for rollups vs. calldata.\n- Key Benefit 2: Lays foundation for full Danksharding and ~1.3 MB/s data bandwidth.
~100x
Cheaper Data
1.3 MB/s
Future Bandwidth
03
The Pivot: Ethereum as the Sovereign Data Layer
The endgame is a modular stack where Ethereum provides bulletproof consensus and data availability, while execution migrates to specialized layers. This turns Ethereum into the universal settlement and security hub, anchoring L2s, validiums, and projects like Celestia-inspired alt-DA.\n- Key Benefit 1: Unlocks horizontal scalability without fracturing liquidity or security.\n- Key Benefit 2: Creates a vibrant data market, where value accrues to ETH stakers for securing the base layer.
Unlimited
Execution Scale
$100B+
Secured Value
DATA AVAILABILITY AS A FIRST-CLASS RESOURCE
DA Layer Competitive Landscape: Ethereum vs. Alternatives
A comparison of how leading DA layers treat, price, and secure data, which is the foundational resource for rollup scaling.
| Feature / Metric | Ethereum (EIP-4844 Blobs) | Celestia | EigenDA | Avail |
|---|---|---|---|---|
Data Pricing Model | Per-blob gas auction, separate from execution | Pay-for-blockspace, fixed per-byte fee | Proof-of-Stake stakers rent out bandwidth | Pay-for-blockspace, fixed per-byte fee |
Data Persistence Guarantee | Permanent (via Ethereum consensus) | ~2 weeks (Data Availability Sampling window) | 21 days (EigenLayer operator slashing window) | Permanent (via Substrate/Polkadot consensus) |
Throughput (MB/s) | ~0.75 MB/s (post-Dencun, ~6 blobs/block) | Up to ~100 MB/s (theoretical, with 1MB blocks) | Up to ~10 MB/s (target for Phase 2) | Up to ~6.7 MB/s (with 2MB blocks) |
Cost per MB (Current, USD) | $0.10 - $1.50 (volatile, gas-based) | $0.001 - $0.01 (stable, fee-based) | < $0.001 (subsidized, cost-plus) | $0.001 - $0.01 (stable, fee-based) |
Security Model | Ethereum L1 Validators (~$110B stake) | Celestia Light Clients (Data Availability Sampling) | EigenLayer Actively Validated Services (~$20B restaked) | Polkadot Shared Security (via parachain) |
Settlement Integration | Native (Rollups settle directly to L1) | Bridging required (e.g., to Ethereum L1) | Bridging required (to Ethereum L1 for settlement) | Bridging required (to Ethereum or other L1) |
Data Pruning Post-Window | ||||
Proposer-Builder Separation (PBS) |
deep-dive
THE COST OF TRUST
First Principles: Why Native DA Beats Outsourcing
Ethereum's native data availability is a non-negotiable primitive that external solutions cannot replicate without introducing systemic risk.
Native DA is finality. Data posted to Ethereum L1 is secured by the full consensus of validators, making it the single source of truth for all L2s like Arbitrum and Optimism. Outsourced DA from Celestia or Avail creates a separate trust assumption, fragmenting security.
Outsourcing creates reorg risk. A malicious sequencer with external DA can withhold data, preventing fraud proofs and freezing funds. This is the core vulnerability that EigenDA and other restaking-based systems attempt to mitigate with economic slashing.
Ethereum treats data as a first-class resource because execution is cheap to verify but expensive to compute. Blobs from EIP-4844 encode this principle, providing a cheap, high-throughput lane specifically for verification data, not general computation.
Evidence: The total value secured (TVS) on Ethereum L2s exceeds $40B. This capital demands the cryptoeconomic finality of Ethereum's base layer, not the probabilistic security of an external data chain.
counter-argument
THE DATA
Steelman: The Case for External DA Layers
Ethereum's core innovation is its robust, verifiable data layer, which rollups now commoditize by sourcing cheaper data elsewhere.
Data is the constraint. Ethereum's security model depends on global state verification, which requires every node to process every transaction's data. This creates a fundamental scalability bottleneck that limits throughput and inflates costs for all applications.
Rollups separate execution from data. Layer 2 solutions like Arbitrum and Optimism perform computation off-chain but historically posted all transaction data back to Ethereum. This design treats Ethereum as a high-cost data ledger, paying for security they do not fully utilize.
External DA is a logical optimization. Protocols like Celestia and EigenDA provide specialized data availability at a fraction of Ethereum's cost. Rollups like Manta Pacific and Mantle use them, proving the market demands cost-efficient data verification separate from execution security.
Evidence: Posting 1 MB of calldata to Ethereum costs ~$800; the same data on Celestia costs ~$0.01. This 1000x+ cost differential forces rollups to optimize for data, not just execution.
takeaways
THE DATA-CENTRIC LAYER-1
TL;DR for Protocol Architects
Ethereum's core innovation isn't just smart contracts; it's making verifiable data a primary, constrained resource, fundamentally shaping its security and economic model.
01
The Problem: Gas is a Data Tax
Gas fees aren't just for compute; they're a market mechanism to price and constrain on-chain data storage and transmission. Every opcode and byte of calldata has a cost, making data the ultimate scarce resource.
- Key Benefit 1: Prevents state bloat and denial-of-service attacks by economically disincentivizing spam.
- Key Benefit 2: Creates a clear, predictable cost model for protocol design (e.g., ~16 gas per non-zero byte of calldata).
~16 gas
Per Calldata Byte
~1.5MB
Block Data Limit
02
The Solution: EIP-4844 & Proto-Danksharding
Scaling data availability (DA) without compromising security. Introduces blob-carrying transactions—large data packets (~128KB each) priced separately and deleted after ~18 days.
- Key Benefit 1: Drives ~10-100x cost reduction for L2 rollups (Optimism, Arbitrum, zkSync) by decoupling data publishing from execution gas.
- Key Benefit 2: Preserves Ethereum's security as the canonical DA layer, preventing fragmented security models like alternative DA from Celestia or EigenDA.
~128KB
Per Blob
-90%
L2 Cost Target
03
The Architectural Imperative: State is Sacred
Ethereum's global state is the single source of truth, but accessing it (SLOAD) or changing it (SSTORE) is intentionally expensive. This forces protocols to optimize for state minimalism.
- Key Benefit 1: Encourages designs like Uniswap V3's concentrated liquidity, which packs more efficiency into fewer storage slots.
- Key Benefit 2: Makes statelessness and state expiry (via Verkle Trees) viable long-term scaling paths, keeping nodes lightweight.
20k gas
SSTORE Cost
~1 TB
Full Node State
04
The Consequence: The L2-Centric Future
High on-chain data costs made rollups inevitable. Ethereum as a settlement and DA layer is the direct result of treating data as first-class. Execution migrates to L2s, which compete on efficiency.
- Key Benefit 1: Enables specialized execution environments (e.g., StarkNet for ZK, Arbitrum for general EVM) while anchoring security to Ethereum.
- Key Benefit 2: Creates a clear value accrual stack: ETH pays for global security and data, while L2 tokens (if any) capture execution value.
$30B+
L2 TVL
>90%
Txn Share on L2
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