Why does Based Rollup need pre-confirmation technology?

Author: LindaBell Source: chainfeeds

2020 In October of this year, Ethereum Lianchuang Vitalik released the "Rollup-centered Ethereum Roadmap." As of now, we are in a multi-Rollup ecosystem. While the Ethereum Rollup roadmap brings significant benefits such as improved user experience, it also raises a new set of challenges. For example, developers often need to deal with fragmentation and decision-making dilemmas when facing a diverse Rollup ecosystem. In addition, centralized sorters also bring issues such as transaction review and MEV utilization, causing some users to face unfair costs or operating experiences.

In order to solve the problem of centralized sorter, the current mainstream solutions include shared sorter and Based Rollup. Although the shared sorter can provide unified sorting services for multiple Layer 2, as a third party, there are issues with trust and incentive mechanisms. Based Rollup improves decentralization by relying on the proposer ordering of Ethereum L1, but its transaction confirmation time depends on the block time of Ethereum L1 (about 12 seconds) and cannot achieve fast confirmation.

The historical origin of preconfirmation

Preconfirmations (or "Preconfs") can be understood as "providing credible notices before formal confirmation". To give a simple example, when you call to make a reservation at a restaurant, the restaurant confirms your reservation. This is actually a kind of pre-confirmation behavior: you haven't arrived at the restaurant yet, but you have already received a credible commitment to ensure that you have a place to eat at the designated time, so you can save time in line and do other things.

In fact, pre-confirmation is not a new concept. As early as the early days of Bitcoin's development, in order to make Bitcoin easier to use, the Bitcoin community proposed the concept of "0conf", which means that before the Bitcoin network fully confirms the transaction, the transaction can be Be considered valid, thereby reducing the wait time for final confirmation of the transaction. In 2023, Uri Klarman introduced this idea into Ethereum and proposed the concept of “Chained Preconfirmations”. The core idea is that future pre-confirmors can not only pre-confirm the current transaction, but also inherit the pre-confirmations already made by all previous pre-confirmors. Primev then further explored this concept and proposed a new bidding mechanism "pre-conf" to allow the block builder alliance to commit to the most effective way to include a certain transaction within a specific period of time. Allow users to participate in pre-confirmed bidding.

(Primev  is building mev-commit. mev-commit is a decentralized platform specifically designed to coordinate MEV transactions. Through real-time encryption commitment and efficient execution bidding mechanism, it helps users and providers ensure the reliability and speed of transactions, suitable for complex trading scenarios that require fast confirmation and execution, such as pre-confirmation and other high-frequency trading strategies). >

Justin Drake further promoted the development of this concept and proposed the concept of "Based preconfirmations". Currently, L2 mostly uses centralized sequencers. Its biggest advantage is that it can provide users with pre-confirmation. This confirmation means that the sequencer has promised to include user transactions in future blocks. However, this confirmation is not final and users may still experience transactions being reordered or canceled. The "Based preconfirmations" mechanism proposed by Justin Drake aims to provide instant preconfirmation while allowing L1 to complete decentralized sorting (Based Rollup).

Proposal and optimization of the Based preconfirmations mechanism

The concept of Based Rollup was first proposed by Vitalik Proposed in the article "An Incomplete Guide to Rollups" published in 2021. In this article, Vitalik introduces the concept of "Total Anarchy", describing a completely decentralized environment in which anyone can submit at any time Transaction batch (batch), without any restrictions.

In March 2023, Justin Drake published the article "Based rollups—superpowers from L1 sequencing", further clarifying the concept of Based Rollup. When the ordering of a rollup is driven by L1, it is called a Based Rollup or an L1-ordered Rollup. Under this architecture, the generation and ordering of Rollup blocks is carried out through proposers on L1. L1 proposers can cooperate with searchers and builders to include Rollup blocks directly in L1 blocks. middle. It is worth noting that Based Rollup blocks are built by L2 builders by default. Therefore, Based Rollup does not increase the burden on L1 verifiers.

By transferring the responsibility for transaction ordering to L1, Based Rollup inherits the safety and liveness of L1 while optimizing the performance of L2. This mechanism brings many advantages, such as reducing L2 infrastructure and operation and maintenance costs. However, there are also some challenges, such asmost of the MEV benefits of Based Rollup will flow to L1. But this does not actually pose a problem, because the main revenue of centralized Rollup comes from L2 transaction fees and MEV opportunities. The former is the main source of revenue, and the latter requires a lot of infrastructure and high maintenance costs. Based Rollup retains the primary revenue source of L2 transaction fees while outsourcing the proposer role to L1 builders, thereby reducing infrastructure and operational costs. Another challenge is that Based Rollup cannot achieve fast confirmation of transactions. Since Based Rollup's transaction confirmation time is entirely dependent on Ethereum L1's block time (currently around 12 seconds), this is in sharp contrast to the fast soft confirmation provided by a centralized sequencer.

In order to solve the problem of transaction confirmation speed in Based Rollup, Justin Drake proposed "Based preconfirmations" in November 2023, executed through Slash Provide pre-confirmationfor Based Rollup. In the original idea, Justin Drake put forward two key premises. One is that the L1 proposer must choose to add additional penalty conditions to become a pre-confirmor (heavy pledge mechanism), and the other is that the L1 proposer must be able to force the transaction to be included. On the chain (inclusion lists).

Subsequently, Justin Drake optimized this idea, allowing Rollup to use Ethereum L1 for shared sorting and pre-confirmation, which is simpler and does not require a hard fork. In the new proposal, some validators voluntarily become sequencers, and validators who do not want to sequence become "includers". They can include transactions but not sequence them. And the sorter can reorder the transactions of includers and even insert more transactions into them. Users can choose two ways to submit transactions: one is to send the transaction to the backup memory pool and pay a small fee to have includers include their transaction; the other is to communicate with the next orderer and pay a higher pre-confirmation fee to request pre-confirmation. If the orderer fails to settle transactions within specified slots, includers are still able to process those transactions.

Compared with the original proposal, the optimization plan directly leverages the economic security of Ethereum L1, allowing L1 proposers to voluntarily become orderers without Additional penalties reduce reliance on additional infrastructure.

From concept To practice: Application of Based preconfirmations

Taiko: Pioneer in developing and promoting Based Rollup

Since the second half of 2023, Taiko has been developing towards becoming a Based Rollup. As Based Rollup, Taiko relies on a decentralized system of proposers running specific clients (taiko-geth) and synchronized with an L2 memory pool. When the proposer finds that a batch of transactions can form a profitable block, the proposer will submit the packaged block to Ethereum L1.

During transaction processing, Transactions initiated by users on Taiko will first enter the L2 memory pool. L2 searchers look for profitable transactions in the L2 memory pool and package them into L2 transaction packages. Subsequently, the L1 searcher acts as the L2 block proposer, sorts the transactions packaged by the L2 searcher into an L2 block, and then includes the block in its L1 transaction package and integrates it into the L1 block. It is finally verified by L1 validators and submitted to the Ethereum main chain to ensure the finality of the transaction.

In order to improve efficiency, Taiko has introduced a pre-confirmation mechanism. Proposers can publish pre-confirmations to other participants in the network before officially submitting a block, telling them which transactions will be included in the upcoming block. In addition, the proposer can also periodically publish sequentially ordered small batches of pre-confirmations to the block builder, and the builder can choose to publish these pre-confirmed transactions to L1 in batches, rather than publishing a large batch at once. data blocks, thereby reducing the cost of data release.

In addition, in a decentralized proposer system, there may be situations where multiple proposers try to submit blocks containing the same transaction at the same time. If this happens, only one of the blocks will be accepted by L1, while the other blocks will be rolled back, and the failed proposer will lose the block fee. To avoid this potential conflict, Taiko introduces a leader election mechanism. At any given time, only one proposer is elected as the leader and has exclusive rights to finalize blocks. This ensures that blocks created by the leader are added to the blockchain while blocks from other proposers are deprecated, preventing multiple conflicting blocks and avoiding failed proposals. The person loses the fee for submitting the block.

Puffer UniFi: Committed to reshaping the decentralized application chain ecosystem

Puffer Finance also Launched the Puffer UniFi solution based on the pre-confirmation mechanism. As Based Rollup, transaction ordering on UniFi is outsourced to Ethereum L1, while Puffer UniFi’s pre-confirmation mechanism is implemented through Puffer’s native heavy-staking validator.

In terms of the specific process, the user submits Transactions are first processed by Puffer validators, and these validators have been registered as "Native Restaking" nodes on Ethereum. Validators provide pre-confirmation commitments to users in about 100 milliseconds, letting users quickly know that their transactions have been received and will be included in future blocks. To ensure that validators will adhere to their pre-confirmation commitments, Puffer is also deploying the UniFi AVS mechanism to impose additional penalty conditions on validators. After providing pre-confirmation, Puffer validator validators will package these transactions with other transactions and submit the block to Ethereum L1. Finally, the Puffer UniFi smart contract Puffer Sequencer Contract accepts batch transactions, ensuring that the transaction status has been confirmed and cannot be rolled back.

According to the latest roadmap, Puffer's UniFi test network will be launched in September 2024 and UniFi AVS The mechanism is expected to go online with the UniFi mainnet in The fourth quarter of 2024.

Puffer UniFi’s vision is to solve the current problem of liquidity fragmentation between chains. In the future, applications built on Puffer UniFi will be regarded as independent application chains. These application chains rely on the L1 sorting and pre-confirmation mechanism provided by UniFi, which can not only reduce development costs, but also integrate with other rollups or application chains based on L1 sorting. Seamless interoperability. For example, user Alice wants to use her tokens on Rollup A as collateral to borrow assets on Rollup B. Since both Rollup A and Rollup B rely on the same Ethereum L1 validator for transaction ordering, this validator can Alice's transactions on two Rollups are processed simultaneously in the same Ethereum block. In this way, Alice can operate between different Rollups without the need for complex asset transfers through cross-chain bridges.

What else deserves attention?

In addition to the two projects mentioned above, there are also some pre-confirmation related content worthy of attention. For example, Primev launched the mev-commit platform. Primev can provide pre-confirmation service for any Based Rollup through mev-commit, and proposes a blob pre-confirmation solution that combines mev-commit and inclusion list (IL); Espresso proposes a Proposer-promised pre-confirmation solution, each Rollup can customize the sort order and slashing conditions of Proposer-promised pre-confirmations; Spire launched the Preconfirmation Registry, a system for proposers (such as large operators and independent pledgers) to issue collateral in ETH; Chainbound announced Bolt is a protocol that allows Ethereum proposers to commit to the content of their blocks.

In addition, in order to simplify the user experience and better coordinate pre-confirmation requests, researchers are also exploring the introduction of pre-confirmation gateways to reduce the complexity of pre-confirmation Abstract from the user. Through the gateway, proposers can delegate their pre-confirmation rights to the gateway. The gateway is responsible for performing more complex tasks such as communicating with users, maintaining full node uptime, etc. Pre-confirmed Gateway projects to watch include Aestus, Titan and Ultra Sound.

Faced with the liquidity fragmentation problem in the current Rollup ecosystem, Based Rollup is widely regarded as an effective potential solution. For example, Cinneamhain Ventures partner Adam Cochran tweeted that Based Rollups are a potential solution to Ethereum’s Layer 2 economic problems. Pre-confirmation plays a crucial role in improving the user experience of Based Rollup. As some major agreements gradually make positive progress, we will continue to pay attention to the latest developments in this field.