Key Takeaways

1. Introduction

Undoubtedly, the period from 2022 to 2023 marked the ascendancy of Ethereum’s Layer 2, during which various rollup networks proliferated. Within the Optimistic rollup sector, Arbitrum and Optimism have firmly secured spots among the top 10 by total value locked (TVL). Meanwhile, in the area of zk rollups, the second wave of Layer 2 is coming, especially with the introduction of Polygon’s zkEVM and the launch of zkSync Era’s mainnet.

1.1 The various risks of rollups

(Data as of Apr 15, 2023 | Source: l2beat)

However, due to the rapid growth of various Ethereum layer 2 networks, not all of them have met the criteria for a proper rollup. The table above analyzes the risks of existing rollup networks, as provided by the l2beat. Each item describes the approach taken by the rollup network, with yellow indicating a slight risk and red indicating a major risk.

1.2 The Importance of Sequencers

So far, we’ve covered a variety of risks, but one that’s become a hot topic as many rollups have gotten larger is the failure of sequencers and validators. Optimism Network has grown to be the 6th largest network on TVL, but users’ funds can be frozen if a whitelisted validator goes offline, and in the case of the recently released zkSync Era, there is no way to deal with a sequencer and validator going offline.

The reason for this concern is that most rollup networks in existence operate a centralized sequencer. Let’s take a look at how the major rollup networks operate their sequencers:

As a result of the centralization of sequencers and validators, there have been a number of recent projects that have attempted to decentralize the sequencing of rollup networks. Examples include Astria, Espresso, Radius, OP Stack, and Suave. They provide their own sequencer network to the rollup network so that the nodes in the rollup network only need to compute, but they decide the order of the transactions and hand them off to the rollup network. Contrarily, a different approach has been proposed by Ethereum Foundation researcher Justin Drake called Based rollups.

2. Based Rollups

“Base can achieve tokenless decentralization by becoming based.” — Justin Drake

That’s the last sentence of Ethereum Foundation researcher Justin Drake’s introduction to Based Rollups. What does this mean that the rollup network launched by Coinbase could utilize a form of Based Rollups to achieve tokenless decentralization? (Note that Based Rollups and Coinbase’s Base are not related, except for the similarity in name).

2.1 Overview

Based Rollups (or L1-sequenced Rollups) mean that the sequencing of the rollup network takes place on the L1 that the rollup is based on (in most cases, the Ethereum network). More specifically, in the case of Ethereum, this means that searchers, builders, and proposers on the network all participate in the sequencing of the rollup network.

2.2 Participants of Block Building

(MEV-Boost | Source: Flashbots)

Currently, the majority of Ethereum network blocks are created through a middleware called MEV-Boost, and in the future, this process will be built into the Ethereum protocol itself in a Proposer-Builder Separation (PBS). In this approach, searchers initially listen to the mempool, identify MEV opportunities, bundle them, and submit them to the builders. Builders take the transactions in the mempool and the bundles they receive from searchers and use their own algorithms to create full blocks that maximize MEV revenue, which are then delivered to proposers and added to the Ethereum network.

In this process, searchers and builders bid nearly as much MEV revenue as they can generate to get their bundles and transactions selected, with the effect that MEV revenue flows from searchers — builders — proposers.

2.3 Advantages of Based Rollups

Based Rollups have a number of advantages over traditional rollup networks that handle sequencing on their own. First, they rely on Ethereum to sequence transactions, so they benefit from Ethereum’s liveness. When we looked at the risks of different rollups above, we saw that a lot can go wrong if the sequencer or validator fails, but with Based Rollups, that risk doesn’t exist unless something happens to the Ethereum network.

The second is decentralization. Based rollups can still leverage the diverse set of participants that create blocks on the Ethereum network, and anyone can join the sequencing permissionlessly when they see an MEV opportunity in Based Rollups’ mempool.

Third is the economic alignment with Ethereum. If an MEV opportunity is identified in Based Rollups, searchers and builders on the Ethereum network will submit corresponding bids to sequence it, so the value from Based Rollups’ MEV revenue will naturally flow into Ethereum L1.

Fourth is simplicity. A rollup network in the form of Based Rollups is also simpler than having a centralized sequencer. Based Rollups doesn’t need to validate the signatures of sequencers, build escape hatch (a system that allows users to withdraw if a validator goes offline safely), or introduce new algorithms for decentralized sequencers later.

Fifth, no tokens are required. Traditional rollup networks require the presence of a token in order to decentralize the sequencer later. However, since the sequencing of Based Rollups is done on the Ethereum network, no token is required.

2.4 Disadvantages of Based Rollups

Based Rollups also have a number of disadvantages. The first is that the rollup network’s own MEV revenue flows to Ethereum L1 and does not accumulate value on L2. However, in addition to MEV revenue, base fees for submitting transactions can be accumulated at the L2 end.

Second, there are sequencing limitations. Since Based Rollups relies entirely on the Ethereum network for its sequencing process to order transactions, it is very difficult to introduce their own sequencing process. Furthermore, relying on Ethereum for sequencing has the disadvantage that it follows Ethereum’s finality, which can be very UX-unfriendly, but many researchers are currently researching ways to impose finality in advance through EigenLayer, inclusion lists, builder bonds, etc.

3. Example

Although it’s not certain whether Based Rollups will be applied, there is a research piece posted on ethresearch by Taiko, one of the prominent Layer 2 zkEVM projects.

(Source: Taiko)

If L2 applies Based Rollups, the MEV flows like this:

Taken together, if L1 Searchers are watching both L2 and L1 mempools, they can extract cross-chain MEV between Ethereum and L2 with Based Rollups, and the value of L2’s MEV can be accumulated in Ethereum.

4. Final Thoughts

As the various rollup networks grow, there is a growing concern about centralized sequencers. While there are a number of sequencing layer projects out there to address this, Justin Drake’s Based Rollups is a refreshing idea.

I’ve already mentioned the drawbacks above, but in my opinion, the biggest weaknesses of Based Rollups are profitability and slow finality. The only incentive for Ethereum network block producers to participate in the sequencing of Based Rollups is the MEV revenue which might be not lucrative. Also, since the sequencing takes place on the Ethereum network, instead of benefiting from Ethereum’s liveness, they are locked into slow finality. If the sequencer part, which currently has centralization issues, is also handled by Ethereum, it can be said to be closer to a true rollup, but incentive schemes and ways to preemptively grant completeness will need to be further researched before Based Rollups can be widely used.

Coinbase, the largest cryptocurrency exchange in the United States, has launched its own rollup network, Base, with two design goals: 1) tokenless and 2) decentralized. Base published a roadmap to achieve this by leveraging Optimism’s OP Stack, but both goals could have been easily achieved by adopting the similarly named Based Rollups approach. If you’re an enterprise building a blockchain network, and you’re looking for high scalability and decentralization, and you can’t mint tokens, Based Rollups could be a good option.

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links