How does rollup work? What projects are there?

Author: Jun, Bankless; Compiler: Deng Tong, Golden Finance

Ethereum’s scaling has been one of the most discussed challenges in the cryptocurrency space. To address this, the community has chosen a rollup-centric approach. The idea is simple: Instead of hosting all applications on Ethereum, focus on rollups that provide faster, cheaper transactions while still returning to Ethereum. So you get faster, cheaper transactions but still maintain the security of Ethereum.

We now have optimistic rollups like Arbitrum, OP Mainnet, and Base, as well as zk-rollups like ZKsync, all of which have contributed to Ethereum’s growth. They have brought more users and value to Ethereum, however, people are still concerned about their reliance on centralized sorters and the fragmentation of cross-chain liquidity.

This is where rollups come in. They are more tightly integrated with Ethereum’s infrastructure and help retain value in the ecosystem. This new way of building rollups could bring additional value to Ethereum and ETH.

What are rollup-based?

A rollup-based or L1-ordered rollup is a rollup in which the underlying L1 chain (like Ethereum) directly manages transaction ordering.

Unlike traditional rollups that rely on their own orderers, rollup-based rollups leverage the security, liveness, and decentralization of L1 by outsourcing transaction ordering to L1’s infrastructure. This infrastructure includes proposers, searchers, builders, and other participants who permissionlessly include rollup-based blocks in L1 blocks.

Initially, this approach seemed inefficient — Vitalik called it “total anarchy” in his 2021 post on rollups. But today rollup-based rollups have become more viable. In 2023, Ethereum researcher Justin Drake brought the concept back into the spotlight, arguing that rollup-based rollups are not only more consistent with Ethereum, but also eliminate the need to make separate security assumptions for each new rollup.

How do rollups work?

Rollup-based rollups use L1 for consensus, data availability, and settlement layers, while handling execution independently. For example, when Ethereum is the base L1, the key layers of a rollup-based rollup are as follows:

• Execution layer — managed by the rollup itself, transactions are executed off-chain.

• Consensus layer — relies on Ethereum validators to order transactions.

• Data availability layer — uses Ethereum as the DA layer to ensure that anyone can verify transaction data.

• The settlement layer — also part of Ethereum, records the final transaction status of the rollup.

Rollup-based rollups use Ethereum for everything from transaction ordering to settlement. While this approach may not seem much different from traditional rollups, it fundamentally changes the way ordering is handled. Instead of relying entirely on a separate sorter, rollup-based rollups leverage Ethereum itself for transaction ordering.

In a traditional rollup, users send their transactions to a dedicated sorter — essentially a machine operated by the rollup team. The sorter is responsible for collecting user transactions, determining their order, and packaging them into blocks published on Ethereum.

In contrast, rollup-based rollups direct user transactions to the block builders who manage both Ethereum and the rollup. This use of Ethereum infrastructure allows rollups to benefit from the same guarantees provided by Ethereum, allowing transactions to achieve finality more reliably than non-rollup-based rollups.

Top Rollup-Based Projects

As a relatively new concept, rollups are still under development, with a few key projects emerging:

• Taiko — Ethereum equivalent (type 1) ZK-EVM that is maximally compatible with Ethereum and does not introduce additional trust assumptions.

• KeySpace — Coinbase’s zk-rollup, designed to create smart wallets that can be used on any chain.

• Taiko Gwyneth — A rollup using preconfirmations, designed to sync with Ethereum.

• UniFi — A rollup developed by the PufferFi team that uses preconfirmations to enhance the user experience.

• Spire Labs — A framework on Ethereum that allows developers to build blockchain-based applications.

Why use a rollup?

As Justin Drake outlined in his original post, rollups are worth your attention for the following reasons:

• Inherited liveness and decentralization — One of the main advantages of rollups is their ability to inherit the liveness guarantees of the underlying L1 chain. As long as the L1 is operational, the rollup can also operate.

• Economic alignment with L1 — The economic model of a rollup establishes a mutually beneficial relationship with the L1. Priority fees and MEV from these rollups naturally flow to L1. This synergy not only adds value to the base layer, but also boosts the legitimacy and brand awareness of the rollups themselves, thanks to Ethereum’s community.

• Cost efficiency — Outsourcing ordering to Ethereum reduces development costs for rollups, speeds up time to market, and reduces user costs (especially at scale).

However, rollups also have tradeoffs. They sacrifice some profitability by relying on Ethereum for ordering, missing out on priority fees and MEV. In addition, they face challenges inherent to Ethereum, such as slow block times, which can cause latency issues.

These challenges explain why existing rollups are designed with their own sorters — to provide a fast user experience. While solutions such as pre-confirmed inclusion and execution are being discussed as potential ways to enhance the user experience of rollups, it remains uncertain whether such improvements can be achieved without introducing a trusted third party. The real debate is that these changes could undermine the fundamental simplicity and security that made rollup-based computing attractive in the first place.