Layer2 Sequencer problem Uncertain future for decentralization?

Uncertain Future of Decentralization Addressing Layer 2 Sequencer Issues

Author: Haotian, Crypto Observer Source: X, @tmel0211

Several previous articles have discussed in detail the stagnation of OP-Rollup and ZK-Rollup in solving the decentralized Sequencer issue. Does this mean that decentralized Sequencers are completely unsolvable?

With this question in mind, I conducted research on MetisDAO, a Layer2 project that claims to have built the first decentralized Sequencer. I found that the decentralized technology logic for Sequencers is not difficult, but what is difficult is “power delegation”.

The Sequencer solution uses “hard distributed” technical consensus and “soft distributed” social consensus, which seems reasonable, but it will lead to completely different End Games! Why is that?

• Binance funds take over large amount of sats The biggest winners from ordi to sats make a profit of 260 million RMB.
• ORDI A Game of Attention
• SEC Approaching Deadline Can we usher in the golden age of Bitcoin ETF? (Attached deadline)

On the Ethereum mainnet, users submit transactions to the Mempool, which are collected and broadcasted by Searchers, and finally sorted and packaged into blocks by miner Builders based on Gas fees. This means that the transactions submitted by users may be front-run by MEV, and they need to follow the Nonce queuing mechanism. The Gas fees paid are for the EVM execution costs, making transactions expensive, slow, and prone to sandwich attacks on the mainnet.

By analogy, the Layer2 Sequencer is like a closed Mempool, where users submit transactions to the Sequencer, which is responsible for sorting and consolidating the transactions into a block and submitting it to the mainnet. This makes the transaction sorting on Layer2 entirely determined by the Sequencer, and there is no concept of Nonce queuing. The Gas fees paid are shared among the batch of transactions.

As a result, Layer2 transactions are fast and cheap. In theory, after receiving transactions, the Sequencer will fairly sort them based on transaction Nonce, reception time, Gas prices, or a random algorithm. But due to the excessive power of the Sequencer, there are doubts about whether it will engage in MEV or insert malicious transactions.

Currently, the mainstream approach in the Rollup market is to default that the Sequencer is centralized but still “optimistic” solution.

On one hand, as the Sequencer is a core component of Layer2, adopting a distributed Sequencer with potential failure risks at the beginning will affect the user experience. Therefore, most project teams tend to opt for centralized operation in the early stage to ensure security and stability;

On the other hand, when the project develops to a certain degree and has a high demand for processing a large number of transactions, the team gains supreme power. They not only control the Gas pricing but also directly profit from transaction fees. At this point, the market has grown to a certain scale, and they are less willing to promote decentralization and share power. This is the reason why the decentralization process of Sequencers in the four major Layer2 projects is slow.

However, they have all chosen the same Stack strategy architecture independently. This is a strategic plan that opensource the core technology code and attempts to share critical technical components.

By sharing the core Sequencer and creating a Superchain multi-chain combination architecture, they jointly govern through methods such as MPC multi-signatures and governance voting, forming a social “transparent and decentralized” consensus. It’s like the Monkey King appointing several managing seats to his subordinates, but when it comes to major decision-making, everyone still has to rely on the Monkey King, isn’t it?

This type of soft distributed Sequencer solution is suitable for a market of a certain scale, where everyone has absolute trust in the central authority and has desensitized certain trust friction issues (misconduct, challenges), such as the OP Stack.

What is the decentralized Sequencer solution provided by MetisDAO?

Simply put, this is a “hard distributed” solution for building distributed Sequencer nodes.

Multiple Sequencer nodes on Layer2 form a Sequencer pool, and only Sequencers who have pledged 20,000 Metis tokens are eligible to have block generation rights. Users can also choose the appropriate Sequencer node by staking, so that Sequencer block mining can obtain a large number of token incentives, and users staking can also share rewards.

To prevent misconduct by Sequencers, Metis introduces L2 Rangers to sample the blocks and compare and verify them based on their original State status, including whether the order of transactions has been exchanged or whether malicious transactions have been inserted, and so on. This verification behavior itself can also receive mining rewards. Once misconduct by a Sequencer is discovered, the system will slash the assets staked by the misconducting node, and the validating nodes can also receive a portion of the confiscated assets.

Sequencers will join the POS equity queue to obtain incentives, and Verifiers’ mining will also strengthen the sampling and verification work. Through this incentive and punishment model, a decentralized Sequencer system is built.

This “hard distributed” architecture, where Sequencers are operated by multiple entities and have transparent reward and punishment mechanisms, is an effective way to constrain the rights of individual Sequencers and ensure fair ordering.

Finally, both soft distributed and hard distributed are essentially strategic means.

The Stack open-source framework is more suitable for the expansion of Monolithic entities, relying more on intangible trust assets such as market position and brand reputation. The POS decentralized Sequencer is relatively more suitable for modular small entities to overtake on the curve, after all, the most powerful consensus that blockchain has accumulated is the node mining technology consensus model.

In the Layer2 market situation of Rollup As A Service, the Stack framework will reduce the development cost of Layer2 for project parties, but the “trust” issues accumulated by it need to be borne together by all participants.

And the technology consensus of directly hard distributed, on the issue of Sequencer, is more clearly defined, and more suitable for the context of overall slow progress in Layer2 technology and negative public opinion, which can have a more eye-catching starting point and development prospects.

Of course, I have also looked at other decentralized Sequencer technology consensus solutions such as EspressoSys, AstriaOrg, radius_xyz, etc.

And their logic is similar to Metis, making the single Sequencer designed into multiple entities. The only difference is that when it comes to this part of off-chain consensus, Espresso uses Eigenlayer, which extends the verification capabilities of Ethereum Validators based on Restaking; Astria adopts a more modular approach, providing rapid modular Sequencer solutions for Layer2 developers; Radius uses encryption for all transactions entering the Sequencer pool to eliminate misconduct and MEV risks. (I’ll analyze them when I get the chance.)

In short, my point is: The decentralization issue of Sequencer will not stop at the Stack strategy of the Layer2’s four great kings. Although the decentralized solution of technical consensus does have the risk of consensus overload, it has more prospects compared to the MPC multisig governance social consensus.

We will continue to update Blocking; if you have any questions or suggestions, please contact us!