Definition
A blob (Binary Large Object), introduced to Ethereum through EIP-4844 (also known as “Proto-Danksharding”) activated in March 2024, is a new transaction type that allows Layer 2 rollups to attach large data packages to Ethereum blocks at significantly lower cost than using standard calldata. Blobs are temporary data attachments — each blob can hold ~128 KB of data and is automatically pruned from Ethereum nodes after approximately 18 days, making them ideal for rollup transaction data that needs to be verified but not stored permanently on the base layer. Before EIP-4844, Layer 2 networks like Arbitrum, Optimism, and Base had to post their compressed transaction data as Ethereum calldata, which is expensive and permanently stored. Blobs provide a dedicated, cheaper data space that dramatically reduces the cost of Layer 2 data availability, which in turn enables significantly lower transaction fees for L2 users — often by 10-100x reduction in data availability costs.
Origin & History
| Date | Event |
| 2022 | Ethereum roadmap formalized; Danksharding identified as long-term scalability solution |
| Nov 2022 | EIP-4844 (Proto-Danksharding) proposed by Vitalik Buterin, Dankrad Feist, Diederik Loeffen |
| 2023 | EIP-4844 extensively tested on Ethereum testnets; blob client implementations developed |
| Mar 13, 2024 | Dencun upgrade activates on Ethereum mainnet; EIP-4844 goes live |
| Mar 2024 | L2 fees immediately drop 90%+ as rollups begin posting data as blobs |
| Apr 2024 | Blob capacity fills; blob base fees rise; long-term scalability debate continues |
| 2024 | Ethereum roadmap updates focus on full Danksharding as follow-up to blob transactions |
“EIP-4844 reduces L2 fees by 10-100x. This is the most impactful Ethereum scalability upgrade since the Merge.” — Ethereum developer community, March 2024
How It Works
“` BLOB TRANSACTION STRUCTURE (EIP-4844) ========================================
Standard Ethereum Block (pre-EIP-4844): ┌──────────────────────────────────┐ │ Block Header │ │ Transactions │ │ └── L2 data as expensive │ │ calldata (permanent) │ └──────────────────────────────────┘
Post-EIP-4844 Block with Blobs: ┌──────────────────────────────────┐ │ Block Header │ │ Transactions (regular + blob) │ │ Blob Sidecar: │ │ ├── Blob 1: 128KB L2 tx data │ │ ├── Blob 2: 128KB L2 tx data │ │ └── Blob 3: 128KB L2 tx data │ │ (pruned after ~18 days) │ └──────────────────────────────────┘
HOW L2 USES BLOBS: L2 batches user transactions ↓ Posts compressed data to Ethereum as blob ↓ Ethereum validators verify blob is available ↓ Blob data available for fraud proof period (~7 days) ↓ Blob data pruned from nodes (not stored forever) (KZG commitment remains permanently in block header) “`
| Feature | Before EIP-4844 | After EIP-4844 |
| L2 data posting | Calldata (~16 gas/byte) | Blob (~1 gas/byte) |
| Data permanence | Forever | ~18 days |
| Max data per block | ~60KB calldata | ~384KB blobs (3 blobs × 128KB) |
| L2 fee reduction | Baseline | 10-100x cheaper |
| Cryptographic commitment | None separate | KZG polynomial commitment |
In Simple Terms
- Cheap temporary storage for rollups: Before blobs, Layer 2 networks had to store their data permanently on Ethereum’s main chain (expensive). Blobs provide a cheaper storage area that’s only needed temporarily — about 18 days, long enough for fraud proofs.
- A dedicated data lane: Think of Ethereum’s regular block space as an express lane — expensive but permanent. Blobs are a separate “data only” lane that’s cheaper and automatically cleared after use.
- The KZG commitment magic: Even when blob data is deleted, a small cryptographic commitment (called a KZG commitment) stays in the block header forever, allowing anyone to verify the data existed and was correct.
- Why 18 days?: Optimistic rollups have a 7-day fraud proof window. Blobs are kept for 18 days (safely beyond 7 days) to ensure anyone can verify L2 data during the dispute period, then discarded to save storage.
- Massive fee reduction: When Arbitrum and Optimism switched to posting blob data instead of calldata after Dencun, L2 user fees dropped from typically $0.25-$2 per transaction to $0.001-$0.01 — transforming L2 economic viability.
Real-World Examples
| Scenario | Implementation | Outcome |
| Dencun upgrade (Mar 2024) | EIP-4844 activates; L2s switch to blob data posting | L2 fees drop 90-99% immediately; Optimism, Arbitrum, Base fees reach sub-cent |
| Arbitrum blob adoption | Arbitrum One switches transaction data to blobs | Average user fee falls from $0.25 to $0.002 for most transactions |
| Base chain growth | Coinbase’s Base L2 uses blobs; ultra-low fees drive adoption | Base transaction volume increases dramatically; more viable for consumer apps |
| Blob capacity surge | Popular NFT mint fills all blob capacity in a block | Blob base fee spikes; L2 fees temporarily increase demonstrating market mechanism |
| Full Danksharding roadmap | EIP-4844 as “proto” step toward full data sharding | Ethereum roadmap targets 64+ blob target per block with full Danksharding |
Advantages
| Advantage | Description |
| Massive L2 fee reduction | 10-100x cheaper data posting for rollups directly reduces user fees |
| Dedicated data space | Blobs don’t compete with regular transaction gas; separate fee market |
| Temporary storage | Only stored as long as needed; reduces long-term Ethereum node storage burden |
| Cryptographic guarantee | KZG commitments ensure data availability is provable even after pruning |
| Roadmap stepping stone | Proto-Danksharding designed to evolve into full Danksharding for even greater scale |
Disadvantages & Risks
| Disadvantage | Description |
| Capacity limits | Current 3-6 blobs per block (target/max) still limited; can fill during high demand |
| Blob fee volatility | When blobs fill, blob base fee spikes independently of ETH gas price |
| Not permanent | Data pruned after ~18 days; long-term data availability requires separate solutions |
| Complexity | KZG cryptography and blob mechanics add protocol complexity |
| Node requirements | Handling blob data requires more bandwidth and short-term storage from nodes |
Risk Management Tips:
- If using L2s for time-sensitive applications, monitor blob fee markets during congested periods
- For applications requiring permanent data availability, supplementary data storage (IPFS, Arweave, EigenDA) remains necessary
- Understand that blob-based L2 security still depends on fraud proofs and validator availability during the 18-day window
FAQ
Q: What does EIP-4844 stand for?
A: EIP stands for Ethereum Improvement Proposal; 4844 is the proposal number. EIP-4844 is also known as “Proto-Danksharding” — “proto” because it introduces the blob transaction format without the full data sharding mechanism planned for future Danksharding, and “Danksharding” after Dankrad Feist, one of the co-authors.
Q: How much did EIP-4844 reduce L2 fees?
A: After Dencun activated EIP-4844 on March 13, 2024, L2 fees dropped dramatically. Optimism, Arbitrum, and Base all saw 90-99% fee reductions for many transaction types. Typical L2 fees went from $0.25-$2 to $0.001-$0.05 for most operations.
Q: Why are blobs pruned after 18 days?
A: Optimistic rollups need fraud proof data accessible for 7 days (their dispute window). Blobs are kept for ~18 days to safely cover this window. After that, the data is no longer needed for L2 security — it was only needed to allow challenge if a rollup posted invalid data.
Q: What is Danksharding and how does it relate to blobs?
A: EIP-4844 (Proto-Danksharding) introduced the blob transaction format as a stepping stone. Full Danksharding will implement data availability sampling (DAS), allowing nodes to verify blob availability without downloading all blob data, and scale to 64+ blobs per block. It’s the multi-year completion of the scalability roadmap that EIP-4844 began.
Q: Do regular Ethereum users need to do anything because of EIP-4844?
A: No. EIP-4844 is infrastructure-level and transparent to users. L2 networks automatically switched to posting blob data. The only effect users notice is dramatically lower L2 transaction fees — a purely positive change.
UPay Tip: EIP-4844 was the most impactful Ethereum scalability upgrade since the Merge — it reduced Layer 2 fees by 90%+ overnight, making Ethereum L2s genuinely competitive with other blockchains for everyday transactions. If you’ve avoided L2s due to fees, check them again post-Dencun.
Disclaimer: This content is for educational purposes only and does not constitute financial or investment advice. Cryptocurrency investments carry significant risk. Always conduct your own research before making any investment decisions.
UPay — Making Crypto Encyclopedic
