Will Ethereum's History Expiry Give Birth to a New Data Market?
It will, but this new market does not appear out of nowhere. It rests on a fundamental shift: historical data moves from mandatory storage to voluntary storage. Ethereum nodes began implementing EIP-4444 in July 2025, deleting historical data older than one year to save 300–500 GB of disk space. Whoever fills this gap is creating a new data market.
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！
Historical data transforms from a "free public good" into a "cost-bearing commodity"
Before EIP-4444 took effect, all full nodes on the Ethereum network stored every piece of historical data dating back to the genesis block by default. This was not driven by economic incentives but by protocol requirements — node operators shouldered the storage cost out of "altruism".
This model has two problems:
- Ever-rising costs: Before pruning historical data, a Geth node stored around 925 GB total, of which historical data (blocks / transaction receipts) accounted for roughly 628 GB.
- No storage incentives: The protocol offered no mechanism to encourage storage or to penalise deleting historical data.
EIP-4444 changes this default assumption: clients may locally delete historical data older than one year and have no obligation to keep serving it on the P2P layer. Yet historical data remains essential for L2 verification, index construction, audits and other scenarios.
Demand persists while supply is cut off — that is exactly the condition for a market to emerge.
Three signals that a new market is forming
Signal 1: Official endorsement of "out-of-protocol distribution" as legitimate
The design documents for EIP-4444 explicitly list three distribution channels for historical data: institutional providers (voluntary archival hosting), torrent downloads (decentralised voluntary hosting), and the P2P network (relying on the availability of nodes that retain the data). The protocol has actively acknowledged that "historical data must be obtained through off-protocol means". This effectively greenlights commercial data services.
Signal 2: Multiple solutions are already being tested and operated
| Solution | Model | Incentive mechanism | Status |
|---|---|---|---|
| Portal Network | Lightweight P2P access network | No direct incentive (altruistic) | Not yet officially live in production as of 2026 |
| EthStorage | Modular storage network | On-chain contract pays storage fees, released to storage nodes over time | Running on Sepolia testnet |
| Swarm | General-purpose decentralised storage | BZZ token economic model; storage requires purchasing postage stamps | Can be connected to Ethereum historical data |
| Covalent EWM | Light client + archival service | Commercial data service model | Testnet is live |
Although the Portal Network is Ethereum's official historical data solution, it lacks storage incentives — nodes run on a voluntary basis. Schemes with economic incentives, such as EthStorage and Swarm, are more likely to form sustainable markets.
Signal 3: EIP-4444 has already landed in production
On 8 July 2025, EIP-4444 officially took effect, with all major execution clients supporting partial history expiry:
- Geth v1.16.0+: supports the
prune-historycommand and the--history.chain postmergeflag - Nethermind 1.32.2+: history expiry enabled by default
- Besu 25.7.0+: supports offline/online pruning
- Erigon v3.0.12+: supports the
--history-expiryflag
This is a fait accompli, not a future imagination.
Judging whether this market is affecting you
Scenario A: You are an ordinary user (checking balances, making transfers, interacting with dApps).
History expiry will not affect your daily use. Accessing current balances, executing transactions, lending and similar operations are completely unaffected.
Scenario B: You are a developer, data analyst or L2 project team.
You depend on historical data for indexing, L2 verification, auditing or state queries. In the future you may need to obtain data through the following means:
- Pay for third-party data services (e.g., Covalent, The Graph)
- Run a full archive node (higher cost, requiring around 2 TB of disk space)
- Obtain free but possibly incomplete historical data via the Portal Network
What counts as complete: You have confirmed whether your use case relies on historical data older than one year.
Common failure reason
Failure reason: Using a new node's "fast sync" mode to query historical data from the genesis block.
Fast sync (snap sync) only synchronises the latest state and does not download all historical blocks. After EIP-4444 takes effect, new nodes can no longer fully sync history through the P2P network by default. If you need historical data, you must plan ahead to obtain it from other channels.
Risk warnings
- Centralisation risk: If the bulk of historical data demand shifts to a handful of commercial service providers (e.g., Infura, Etherscan), Ethereum's data availability will become dependent on these centralised entities.
- L2 verification risk: In the past, L2 nodes could verify their own state by replaying L1 history. After history expiry, relying on third-party storage introduces additional trust assumptions.
- Storage cost uncertainty: The economic models of decentralised storage solutions are still in early stages. Storing BLOBs via EthStorage requires prepaying storage fees, and costs depend on network supply and demand.
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！
How to confirm completion
If you run an Ethereum node, check the client version and the results of running the command:
- Geth users: after running
geth prune-history, check whether disk space has been reduced by 300–500 GB. - Confirm that node synchronisation remains normal and JSON-RPC responses are correct. History pruning should not affect the verification of the head block.
Next steps: If you rely on historical data for development or analysis, confirm your data access channels before EIP-4444 fully takes effect (during the rolling expiry phase). Specifically:
- Check whether the RPC provider you currently use (e.g., Infura, Alchemy) still offers historical data — most paid plans still retain it.
- If you run your own archive node, confirm that you have adequate disk space (2 TB or more).
- Keep an eye on the progress of decentralised solutions like the Portal Network and EthStorage — if they go live officially, they will be more decentralised alternatives to centralised services.
