Ethereum's Scaling Roadmap: Rollups and Layer 2s

Ethereum has a cost problem, and that cost problem is a sovereignty problem. When a simple token transfer costs $5 in gas fees during calm markets — and $50 or more during congestion — the network’s tools for financial self-reliance become accessible only to those moving large enough sums to absorb the overhead. Vitalik Buterin has framed this as the scaling trilemma: a blockchain can optimize for decentralization, security, and scalability, but not all three simultaneously. Ethereum’s architects chose to prioritize decentralization and security at the base layer and push scalability outward, to a constellation of Layer 2 networks called rollups. Whether this strategy serves sovereignty or fragments it depends on what gets built on top — and how honestly we assess the trade-offs.

The Trilemma and Ethereum’s Bet

The scaling trilemma is not a law of physics; it is an engineering observation about current architectures. A blockchain that processes more transactions per second either requires more powerful nodes (which reduces decentralization, because fewer people can afford to run them), relaxes its validation requirements (which reduces security), or finds a way to move computation off the main chain while preserving its guarantees. Ethereum chose the third path.

The bet is straightforward: Ethereum’s Layer 1 becomes a settlement and data availability layer — the bedrock that other systems anchor to. The actual transaction processing happens on Layer 2 rollups, which execute transactions in their own environment and then post compressed results back to Ethereum. The base layer remains decentralized and secure. The rollups provide throughput. In theory, you get all three properties; in practice, the theory is still catching up to the implementation.

This is a deliberate architectural decision, not a stopgap. Buterin’s rollup-centric roadmap, articulated in blog posts from 2020 onward, reoriented Ethereum’s entire development trajectory. Earlier plans for “sharding” — splitting the base layer into parallel processing chains — were largely abandoned in favor of making Ethereum the best possible foundation for rollups. The implications are significant: if you use Ethereum today, you are increasingly expected to use it through a Layer 2, not directly on the base chain.

How Rollups Work

A rollup executes transactions in its own environment, then compresses the results and posts them to Ethereum. The key insight is that posting data to Ethereum is much cheaper than executing computation on Ethereum. By doing the heavy lifting off-chain and using Ethereum only for finality and data storage, rollups can offer dramatically lower fees — often by a factor of ten to a hundred — while inheriting Ethereum’s security guarantees.

Two families of rollups have emerged, distinguished by how they prove their results are correct.

Optimistic rollups assume that transactions are valid by default and provide a challenge period — typically seven days — during which anyone can submit a fraud proof if they detect an invalid state transition. Arbitrum and Optimism are the two largest optimistic rollups. They have attracted significant adoption, particularly in DeFi, because they are relatively simple to build on and compatible with existing Ethereum tooling. The trade-off is that withdrawal period: moving assets from an optimistic rollup back to Ethereum’s base layer requires waiting for the challenge window to close, which introduces friction that does not exist in traditional financial systems.

ZK-rollups take a different approach. Instead of assuming validity, they generate cryptographic proofs — zero-knowledge proofs — that mathematically demonstrate the correctness of every batch of transactions. There is no challenge period; the proof itself is the verification. zkSync, StarkNet, and Polygon zkEVM are the leading projects in this category. ZK-rollups are technically more elegant but have been slower to mature, because generating zero-knowledge proofs is computationally intensive and the tooling for developers is less mature than the optimistic rollup ecosystem.

EIP-4844, implemented in Ethereum’s Dencun upgrade, introduced “blob” transactions — a new data type specifically designed to reduce the cost of posting rollup data to Ethereum. This is the base layer actively making itself a better foundation for Layer 2s. The reduction in L2 costs has been substantial, bringing transaction fees on major rollups down to fractions of a cent for simple transfers. This matters for sovereignty because it determines the minimum economic threshold at which Ethereum’s tools become practical.

The User Experience, Honestly

The rollup-centric roadmap is technically sound. It is also, in its current state, a fragmented and confusing landscape for ordinary users. This is worth stating plainly, because sovereignty infrastructure that requires expert knowledge to navigate safely is not truly sovereign — it is just a different kind of gatekeeping.

Here is what using Layer 2s looks like today. You hold ETH or tokens on Ethereum’s base layer. You want to use a DeFi protocol on Arbitrum. You use a bridge — a smart contract system that locks your assets on one chain and mints corresponding assets on the other — to move your funds. This bridge introduces risk; bridges have been among the most frequently exploited targets in crypto, with billions of dollars lost to bridge hacks. Once on Arbitrum, your experience is fast and cheap. But your liquidity is now on Arbitrum. If a protocol you want is on Optimism instead, you bridge again. Each bridge is a risk event, and your assets are scattered across multiple networks, each with its own block explorer, its own wallet interface quirks, and its own set of assumptions about security.

Liquidity fragmentation is the practical consequence. In a unified system, all buyers and sellers share one pool. In a multi-rollup world, liquidity is split across L1 and multiple L2s, which means worse prices for trades, more complexity for users, and a general increase in the cognitive load required to manage your own finances. Solutions are being built — cross-rollup messaging protocols, intent-based bridging systems, shared sequencing — but as of this writing, the fragmentation is real and consequential.

The honest assessment is this: the rollup-centric roadmap will likely succeed in making Ethereum’s transaction capacity sufficient for global-scale use. The costs have already dropped dramatically. But the complexity it introduces works against the sovereignty thesis in a specific way. When we argue that people should custody their own assets and interact with financial infrastructure directly, we are asking them to navigate a landscape that requires understanding which Layer 2 their assets are on, which bridges are safe, and how to manage positions across multiple networks. That is not an argument against rollups. It is an argument for continued work on the user experience layer, and for honesty about where we are in that work.

What This Means for Your Sovereignty Stack

If you are building a personal financial infrastructure on Ethereum, the scaling roadmap shapes your decisions in concrete ways. First, you will almost certainly interact with a Layer 2 for most transactions. The base layer is for settlement and high-value operations; daily use happens on rollups. Choosing which L2 to anchor to is a meaningful decision — consider the TVL, the maturity of the fraud proof or ZK proof system, the ecosystem of protocols available, and the track record of the team operating it.

Second, minimize bridging. Every bridge operation is a risk event. Where possible, acquire assets directly on the L2 you intend to use, rather than buying on L1 and bridging over. This reduces your exposure to bridge exploits and simplifies your operational picture.

Third, understand that this landscape is not settled. The rollup ecosystem is evolving rapidly; new L2s launch, existing ones upgrade their proof systems, and the base layer continues to change beneath them. Date-stamp your understanding. What is true about fees, TVL, and security assumptions in March 2026 may not be true in March 2027. This is infrastructure in active construction, not a finished building. Use it with the awareness that the floor plan may change.

Gas fees are a form of financial gatekeeping, and Ethereum’s scaling roadmap is an attempt to remove that gate. The progress is real. The work is not done. For sovereignty purposes, what matters is that the tools are becoming accessible at lower economic thresholds, even as the complexity of navigating them remains higher than it should be. The direction is correct; the arrival is not yet.

This article is part of the Ethereum & Smart Contracts series at SovereignCML. Related reading: Ethereum’s Architecture: How It Differs from Bitcoin, DeFi on Ethereum: What’s Production-Ready, Tokens, Standards, and the ERC-20 Economy