Naar hoofdinhoud
Live markt
Block #9

Dossiers

Ethereum and layer 2 dossier

Ethereum has undergone a fundamental shift in recent years: rather than making the base layer itself ever larger, the network moves transactions to layer 2 networks that bundle transactions and record proofs of them on Ethereum. This dossier tracks that strategy, from The Merge to the recent upgrades that drastically cut rollup costs, and looks ahead to the steps still needed before Ethereum can handle the scale its roadmap envisions.

Lieke van DijkWritten by Techredacteur, UtrechtUpdated Checked by the editorial desk

Timeline

  1. The Merge: Ethereum switches from proof-of-work to proof-of-stake.
  2. Dencun upgrade introduces blobs and sharply lowers rollup costs.
  3. Pectra upgrade raises validator limits and improves account abstraction.
  4. Ethereum core developers lock in roadmap for full danksharding.

Why Ethereum isn't simply made bigger

A blockchain that wants to process more transactions per second itself usually has to raise the requirements for nodes: more computing power, more storage, faster network connections. That pushes smaller participants out of the network and concentrates control among a limited number of large parties, which over time undermines censorship resistance and decentralisation.

Ethereum therefore chose a layered approach: the base layer deliberately keeps throughput limited so it remains runnable on relatively modest hardware, while specialised layer 2 networks take over actual transaction processing. That separates the problem of security and decentralisation from the problem of throughput.

This choice is not without criticism. Critics point out that user experience becomes fragmented across dozens of networks with their own liquidity, bridges and risks. Supporters see that as a temporary growing pain that will be solved by better interoperability standards.

Rollups: optimistic versus zero-knowledge

A rollup executes transactions outside the Ethereum base layer but publishes the data and a proof of correctness back to Ethereum. That means the rollup inherits the base layer's security without every transaction needing to be individually recomputed by all Ethereum nodes.

Optimistic rollups assume by default that submitted transactions are correct and give participants a challenge period to prove fraud. Zero-knowledge rollups instead provide a cryptographic proof that the outcome is correct, enabling faster settlement but requiring more complex engineering.

Both families have by now produced large networks with their own user bases and applications, from trading platforms to payment apps. The expectation within the developer community is that zero-knowledge technology will eventually dominate, since it requires no waiting period for withdrawals.

  • Optimistic rollups: fast execution, longer withdrawal time due to the challenge period
  • Zero-knowledge rollups: cryptographic proof, shorter withdrawal time, more complex engineering
  • Both publish transaction data on Ethereum for security and availability

Dencun and the arrival of blobs

The Dencun upgrade of March 2024 added a new data type called blobs, specifically designed for rollups to post their transaction data on Ethereum more cheaply. Before that upgrade, rollups paid for expensive permanent storage space, even though they only need temporary data availability to be able to prove fraud or incorrect computation.

The effect was immediately measurable: transaction fees on major rollups fell by tens of percent to well over ninety percent compared with the period before. That made small, everyday transactions on layer 2 structurally affordable for a broad audience for the first time.

Pectra and the road to full danksharding

The Pectra upgrade of May 2025 built on this with an increase in the number of blobs per block and improvements to account abstraction, letting wallets handle transaction fees and security more flexibly. Validators also gained more room to stake larger amounts without needing to run multiple validator instances.

The ultimate goal is called full danksharding: a design in which individual nodes only need to check small pieces of data availability through sampling, while the network as a whole still guarantees that all data is available. That should multiply capacity for rollups without increasing the requirements placed on individual nodes.

This route is being rolled out in phases, with intermediate steps that gradually raise the number of blobs while developers monitor network stability. Core developers emphasise that each step is only taken once testnets and audits give sufficient confidence.

Bridges, liquidity and fragmentation risk

Moving value between Ethereum and the various layer 2 networks requires bridges. Some bridges are operated by the rollup itself and are relatively trusted, while other independent bridges are offered by third parties and carry an additional layer of security risk of their own.

Bridges have historically proven to be among the most vulnerable parts of the broader Ethereum ecosystem, with several major hacks in previous years. Users are therefore advised to use bridges recommended by the rollup itself and to split large amounts rather than moving them all at once.

  • Native bridges run by the rollup itself are generally considered the most scrutinised
  • Independent bridges often offer speed, but with extra counterparty risk
  • Splitting large amounts across multiple transactions limits damage if something goes wrong

What this means for users and developers

For ordinary users, the layer 2 strategy mainly translates into lower costs and faster settlement for everyday applications, while the base layer is reserved for transactions that require the highest degree of certainty, such as large transfers or settlement of the rollup proofs themselves.

For developers it means that choice of network has become a strategic trade-off between cost, speed, security level and the size of the existing user community on a specific rollup. That fragmentation is partly addressed by cross-chain messaging standards, though this remains an area where the ecosystem is still very much developing.

Frequently asked questions

What is the difference between Ethereum and a layer 2 network?

Ethereum is the base layer that provides security and final settlement. A layer 2 network executes transactions faster and more cheaply and periodically records proof of that on Ethereum.

Why did layer 2 transactions get cheaper after Dencun?

Dencun introduced blobs, a cheaper way for rollups to publish temporary transaction data on Ethereum, sharply reducing the fees rollups pass on to users.

Is money on a layer 2 network as safe as on Ethereum itself?

It depends on the design. Rollups that submit correct proofs to Ethereum inherit much of that security, but bridges and the challenge period of optimistic rollups add extra risks.

What is danksharding?

Danksharding is the intended final phase of Ethereum's scaling strategy, in which nodes check data availability through sampling so that capacity for rollups can grow substantially without heavier requirements on individual nodes.

Do I need to choose which layer 2 network I use myself?

For simple applications, the service you use often determines the network. For larger amounts it is wise to check the age, audits and type of rollup before transferring funds.

Read next

Newsletter

Bitcoin for breakfast, Brussels for lunch

One short email with what actually matters: prices, regulation and the banks that move. No hype, no noise.

  • Daily at 07:00
  • 2 minute read
  • No spam

Free. Unsubscribe in one click.