Layer 2 network’s quest for scalability and cost-effectiveness has led to the wide adoption of the centralized sequencing approach. However, as we know, centralization leads to many on-chain vulnerabilities, including the most common issue of the single point of failure. Now that decentralized sequencers are here and are receiving a good amount of response, Layers2s ought to choose decentralization to overcome the limitations of centralized sequencers and meanwhile improve performance, security, and decentralization. Throughout this article, we will discuss the importance of decentralized sequencers for Layer2 chains. However, we will first highlight the problem caused by centralized sequencers.

Layer2 sequencer centralization concerns

In a centralized setup, Layer2 sequencers are typically run and managed by a single entity, which is often the company that builds the chain. For example, Coinbase runs a sequencer for its Layer2; Base.

Although centralization has its own benefits, like quick settlement and cheap transactions, it often creates issues for many blockchain networks. Also, the idea of centralization itself contradicts the blockchain’s concept of not having a single authority as a network controller.

One incident of single point of failure occurred on the Arbitrum Rollup chain in December 2023, when the sequencer went offline for 78 minutes, causing an unexpected network outage. Although Arbitrum One is the sole rollup chain capable of addressing network issues, it took nearly a week to restore normal operations. This week-long timeframe may not be feasible for hosting a comprehensive global network. Consequently, such events can result in missed opportunities for users and undermine trust in decentralized technology. Like, if a beneficial trading opportunity arose during such a downtime, users would miss out due to the sudden offline status of the singular sequencer managing the entire network ecosystem. This could discourage users from leveraging any Layer2 rollup or appchain environments with centralized sequencers in the future.

Beside common challenges like downtime, centralized sequencer can lead to the following major risks:

• In a centralized sequencer, nodes might be convinced easily to censor particular transactions either by dropping them from the mempool itself or refusing to add those addresses in the first place.

• Centralization of sequencer allows for maximization of extortion out of any specific network, which can be possible through including, excluding, as well as reordering any transactions.

• As continuity in centralized sequencer is not guaranteed, rollups using them will struggle to secure interoperability with other rollup chains.

• Rollups may require additional services from its sequencer network. For example, performing staking with their own custom token, the flexibility to outsource DA (data availability) layer or the sequencer. Having centralization creates issues in enabling all such services.

How can decentralized sequencers change the game?

Decentralization of sequencers refers to random selection of sequencers and the distribution of the transaction sequencing task across them. This means that a single entity does not control decentralized sequencers and selection of sequencers is completely unbiased. Instead, multiple entities across different geographical locations run and manage your chain. This is a critical step for eliminating all the major problems occurring with centralized sequencers that we already discussed earlier in this article.

Think of a decentralized sequencer as a permissionless blockchain, where anyone willing to become a professional validator can join the network and finalize the blocks to get rewarded for their contribution. The benefits are also similar to those of blockchain’s decentralization, which is empowerment of data ownership, enabling truthfulness in the ecosystem, immutability, censorship resistance, resilience, and unhindered robust security.

Further, the adaptable nature of decentralized sequencers encourages innovation, supporting the integration of next-gen features such as the latest concept of shared and dedicated sequencing. Here, a shared decentralized sequencer for Layer2 introduces a concept in which the ecosystem utilizes a common sequencer for the purpose of transaction ordering, aggregating, and committing them to the main chain. Speaking about dedicated sequencers, they are managed through their own network of sequencers that are accountable just to dedicate their service entirely to specific Layer2s without sharing with other chains.

Both the shared and dedicated sequencing have their own benefits and downsides that Laye 2s should consider while making a decision. For example, shared sequencers are better in terms of censorship resistance and it promotes interoperability through pooling all the transactions across interconnected L2 chains. Also, it is an affordable option due to its SaaS like model. But, shared sequencers may not match the performance of dedicated sequencers since it is used simultaneously by multiple L2s. Dedicated sequencers offer higher performance as it uses the network’s resources entirely. Plus, configurations on this model are highly flexible. The potential risks can be breakdown, lesser security and compromise in decentralization.

How Feasible is a Decentralized Sequencer for Layer 2?

From our whole discussion, it is evident that decentralized sequencers are a feasible solution for Layer2 networks because they contribute to robust security, resilience, and decentralization— the core ethos of blockchain. On top of that, decentralized sequencers offer convenient interoperability across Layers, which allows for development of new-age applications such as high-performance and security-rich cross-chain DEXs, multi-chain bridge networks, and futuristic NFT platforms.

Talking about the current scenario, the concept of shared and dedicated sequencing are the preferred options across all the leading rollups. The growing trend of sequencing-as–a-service further gives a push to this. But, as mentioned, even these modes of decentralized sequencing have some downsides. Looking into the solution, we see better alternatives for rollup sequencers that will enable next-level modularity option, robust security, while ensuring decentralization that today’s appchains and rollups seek to achieve. These capabilities will further allow web3 projects and enterprises to launch modular Layer2 chains tailored to cater to specific purposes.

When considering the modularity in rollups, it is important to highlight the integral role of sequencers, especially decentralized ones. If you’ve read our previous article on Cero, you must now have a firm understanding of how a hybrid decentralized sequencer is a game-changing innovation in the blockchain space. Now, for this article, we will discuss the two important feature aspects of Cero: multi-DA and multi-settlement options that are designed to make rollups truly modular for more flexibility and higher performance.

A quick rewind of Cero’s Hybrid Sequencing Network

Cero is a decentralized hybrid sequencing network designed to eliminate all the challenges of centralized sequencing while enabling modularity and high level of security in all kinds of rollups–be it optimistic or zero knowledge. The reason why Cero is called hybrid is due to its provision to allow rollups to opt for shared or dedicated sequencing within a single sequencer network. Below are some of the Cero’s features that makes it a unique, ideal choice for sequencing in modular rollups:

1. Custom staking: Rollups using Cero can perform staking with Cero’s token, or they can use their own native token for this purpose. This means, rollups can implement a custom tokenomics as per their specific validator requirements.

2. Data privacy: Rollups can enable data privacy and accessibility parameters in their ecosystem, and meanwhile they can benefit from the decentralization and security of Layer1.

3. Dedicated sequencing: Rollups can opt for dedicated sequencing to avoid sharing the sequencer with any other chains and hence attaining greater performance.

4. Multi-settlement layer: Rollups can choose any Cero-supported chain to utilize for Layer1 settlement.

5. Multi- DA Layer: Cero offers support to multiple DA layers– both public and private to be used for data availability.

Delving into Multi data availability (DA) and Multi-settlement feature of Cero

As discussed, Cero sequencer is designed in a modular fashion. Hence, it strives to add modularity in rollups, mainly for components that require extensibility the most, such as the data availability and settlement modules. Let’s dive deeper into these multi data availability (DA) and multi-settlement options to better understand their usefulness:

1. Multi data availability (DA) Layer configurability:

Right now, various alternative data availability layers are available for rollups to use, and they are being rapidly adopted due to factors such as excellent data guarantee, greater throughput, low gas fee, and a range of other innovative features. Having understood the significance of alt DAs, Cero provides rollups the choice to use either Layer1 like Ethereum as their data availability layer or they can integrate external public DAs like Celestia and Near DA. That’s because Cero understands that while Ethereum leads in terms of security and ecosystem maturity, alt public DAs are good for scalability & customization.

That was about the rollups that need public DAs, but what the rollup wanting to keep its data confidential? Cero addresses this concern by extending its multi-data availability option to private data storage systems based on data availability committees (DAC) or validium where data is stored and processed off-chain. This configurability option is suited for enterprises and web3 projects that deal with sensitive data and hence do not want to make it public, however they still want to leverage all the benefits of settling on the main Layer1 chain.

All these flexibility aspects makes Cero a preferred decentralized sequencer for all types of rollups, whether its public or permissioned. Also, below is a quick comparison of Cero’s public DA and Private DA that will make this whole Multi data availability (DA) concept even more clear.

1. Multi-settlement Layers configurability:

Similar to data availability, rollups may also seek to use a particular settlement layer. Cero’s chain-agnostic, heterogeneous architecture makes it a novel multi-settlement decentralized hybrid sequencer. This allows rollups to pick a particular blockchain to use for settlement based on their ecosystem–related needs. Rollups can make the selection at the time of their registration on Cero.

As you may already know, the settlement layer in rollups is responsible for settling transactions as well as interactive proofs. Meanwhile, it also accommodates the bridging and liquidity requirements across chains. Each settlement layer (blockchains) comes with its own benefits, for example- some are optimized for cost-effectiveness, or high scalability, or seamless cross-rollup/Layer1 interoperability to attain unified liquidity. Additionally, some rollups want to use layer2s as their settlement layers so that they could be running a Layer3 on top of Layer2. While the requirements can differ, every modular rollup seeks the one main thing, which is fair sequencing.

With Cero, fair sequencing is enabled on all the following settlement chains. Rollups can choose whichever is feasible for them:

• Ethereum

• zkSync Era

• Polygon PoS

• Arbitrum

Also, Cero plans to expand this list of settlement layers with leading blockchains like Polkadot, Optimism, Avalanche, Arbitrum Nova, and few more. Cero has enabled integration of multi-settlement layers, a simplified process with its flexible architecture and soon to be enabled; plugin based integration.

How will this boost modularity in Layer2/layer3 rollups?

Modularity in any rollup refers to its ability to outsource the network’s components like data availability, consensus, or the settlement layer. Cero sequencer enables such modularity in Layer2/Layer3 rollups by providing them the flexibility and efficiency in choosing multi-DA layers and settlement layers for their ecosystem. By integrating a suitable DA or settlement layer, rollups cater to its diverse needs and meanwhile make the whole network operations more simple and manageable. Note that integrating alternative components may require a complex integration approach. However, to compensate for this, modular sequencers offer higher scalability and efficiency with several other benefits.

Wrapping up!

For decentralized sequencers, the modular approach has proved to be a more viable option as more rollups nowadays are prioritizing modularity. Cero utilizes this opportunity by offering multi-DA and multi-sequencer options on its decentralized sequencing network, thereby bolstering rollups modularity. However, rollups on Cero can also proceed with the standard configurations, including for DA and settlement layer. Cero can be a good choice for you if you want to leverage reliable, new-age decentralized sequencer in your rollups, Cero can be a good choice for you. For more information about Cero or any queries, you can schedule a call with our rollups & appchain experts. Or else, mail us all your concerns.

A sequencer is the most vital component in Layer2/Layer3 rollups as it is responsible for transaction ordering– the whole process of collecting transactions, batching them from Layer2 mempool, and then submitting them on the Layer1 chain, e.g., Ethereum. Also, the sequencer resolves fraud proofs received from users.  However, the current sequencing solutions have some issues, such as non-empowerment of custom tokens for staking, increased cost of transaction ordering, slow transaction processing on shared sequencing, and lack of sequencing reconfigurability.

To eliminate all these downsides and make decentralized sequencing to even more mainstream, we are building Cero – a decentralized hybrid sequencing network for public and permissioned rollups optimized to support a wide range of new features, configurations, and security attributes missing today.

The problem with centralized sequencers

Although centralized sequencers are being used by leading Layer1 and Layer2 blockchains, we can’t oversee the challenges that centralization poses related to security, privacy, and usability. For example, in March this year, Polygon zkEVM suffered issues while synchronizing the network’s state, which caused a 14-hour outage. Similarly, Optimism mainnet and Arbitrum also had severe outages due to centralization-based issues in the sequencer.

Here are some of the much-acknowledged problems caused by centralized sequencers:

Transaction censorship: Rollups with centralized sequencers can easily censor any transaction either by dropping them from the mempool or simply refusing to add transactions. However, a decentralized sequencer layer updates the address & other details on the entire Layer2 or on the underlying layer1 so that whenever a particular address is blacklisted, the information is shared with all.

MEV Extraction: Centralization of the sequencer also leads to MEV extraction, where users can include/exclude specific transactions to increase the maximum amount of value they can make from a blockchain.

Single point of failure:  When a rollup is heavily reliant on a single component or entity (in this case, a centralized sequencer), even a simple issue can lead to the collapse of the whole system. This problem can be intentional or accidental.

Inter-rollup interoperability: A centralized sequencer network can be easily optimized to enable cross-rollup interoperability, but due to its inability to ensure continuity, interoperability gets hindered most of the time.

Lack of sequencing configuration: Rollups may want to choose a separate settlement layer instead of Ethereum. Similarly, it may need to store data on a private data availability layer. Centralized sequencers can lead to difficulty in doing all these configurations because of compatibility issues with a range of additional 3rd party integrations.

Staking challenges: A rollup or appchain may want to participate in staking with their own custom token. With a centralized sequencer, modeling your own tokenomics or using custom tokens is not possible, or it can be an extremely complex and resource-intensive process.

Introducing Cero– the hybrid decentralized sequencing network

Cero sequencer is an innovative hybrid decentralized shared sequencing network designed to abstract the current challenges of centralized as well as current decentralized sequencers. With Cero, all kinds of zero-knowledge and optimistic rollups get the flexibility to implement public shared sequencer, public dedicated sequencer, as well as a permissive dedicated sequencer to overcome issues such as transaction censorship, single point of failure, configurability issues, and lack of cross-rollup interoperability. However, Cero goes a step further to implement next-level staking and data privacy features, enabling rollups to stake with their own native token and maintain data confidentiality for them.

How does Cero work?

A decentralized sequencing network can implement its own scalable network to improve the performance of rollups in terms of cost optimization, throughput, and reliability. However, rollups on a decentralized sequencer network cannot achieve higher performance than the sequencer itself. This is a contradiction to rollups’ identity as a feasible scaling solution.

To provide a solution, Cero has unveiled a new concept of hybrid sequencing. The Cero network serves as a shared decentralized sequencer in the first place, but it also has the provision to offer dedicated sequencing to the rollups that demand their own dedicated sequencer for enhanced trust, performance, cross-rollup compatibility, and less complex operations. Also, Cero introduces permissive dedicated sequencing for data-sensitive rollups.

Here, Public shared sequencing is a mode, or we can say it’s a method through which Cero pushes blocks containing their namespaced along with sequenced transactions for all the rollups choosing public shared sequencing and then publishes them on layer1 public blockchain for soft confirmation. The upside of this sequencing model is that it is simple to use, integrate, and manage. However, it is limited in terms of configurations and flexibility. Plus, all the data remains public.

Speaking about Public dedicated sequencing, it works through a random selection of dedicated sequencers from Cero’s sequencer superset. Allotment is done based on the number of dedicated sequencers a rollup needs. Once a rollup registers successfully, the contract on Cero will auto select the sequencers, and allow them to form a close circuit so that they can communicate and reach consensus while maintaining high availability. This mode is preferred for rollups that want greater flexibility and performance while retaining the security benefit of a public sequencer network.

Lastly, there’s Permissive dedicated sequencing: This model is somewhat similar to public dedicated sequencing, but the validator set for rollups choosing Cero’s Permissive dedicated sequencing will not be automatically chosen. Instead, the developer prepares criteria for appointing validators, considering the two conditions below:

• If required, the validators should do additional staking using their own custom token.

• The validator set should be managed by wallet accounts that can vote for on-boarding and off-boarding.

Rollups can either choose one option or both the options at the same time, whichever is suitable.

Also, here’s a graphical representation of how rollups work on Cero’s hybrid sequencing network at the basic level:

1. Rollups (be it a public or private rollup) register on Cero’s decentralized sequencer network.

2. This allows the sequencer network to access rollups’ RPC nodes.

3. RPC nodes then submit transactions directly to the sequencer.

4. Cero stores the data in its ledger as blocks.

5. Next, Cero submits the batched transactions to rollup’s preferred DA Layer and meanwhile publishes only the transaction hash to Layer1. Note that Cero even lets you choose the settlement layer.

Here are some more reasons to choose Cero

As discussed, Cero offers a bunch of novel features that make it an ideal choice for decentralized transaction sequencing. Let’s go into detail of these features:

Customization of staking & incentive strategies:

Cero works by offloading the sequencing task to its PoS network’s participants. These validators have the flexibility to either stake with their rollup’s own token or continue staking with Cero’s native token. Let’s understand this better by looking at different types of rollups and their unique need for staking:

• No additional staking- This staking approach is suited for rollups that do not have their custom token to begin staking. Rather, they want to stake in Cero’s token.

• Additional staking, along with custom token- This approach includes both staking with Cero’s token and staking with rollup’s custom token. Hence, it is ideal for rollups looking for a hybrid staking system.

• Custom token staking- Ideal for rollups that want to stake only in their custom token. As you may already know, this requires a validator set separately from the public network.

Complete privacy of data:

Cero provides rollup chains with the flexibility of achieving data confidentiality with optional privacy. This allows all the rollups registered on Cero to sequence and execute all their transactions privately while inheriting the security from Layer 1. This is particularly useful for blockchains that do not want their sensitive data to be public. Hence, the following two instances are possible with Cero:

1. RPC nodes broadcast all the transactions and the relevant data to the Cero network itself, i.e. public mode.

2. RPC nodes publish all the transaction data to its private data layer or private DAC and meanwhile push only the transaction hashes for sequencing.

Here, the first option is suitable for rollups that are comfortable in sharing data publicly because their data will be published on public Layer1 like Ethereum. Whereas, the second option aims to maintain confidentiality by publishing data on a private DA layer.

Advanced security model:

Cero strives to maintain robust security on its PoS network with its own validator set curated strategically to achieve immutability and decentralization. Cero already boosts security by abstracting the transaction sequencing part from the rollups, then it goes a step further to achieve even more reliable security that does not leave even a single chance of attacks taking place across the entire network. For this, Cero implements the following innovative security mechanisms:

Proposer Builder Separation(PBS)- PBS is the first method that Cero incorporates to separate the entity responsible for building the block containing transactions and the block proposer that is accountable for further submitting these blocks. This segregation is important to ensure that the entity ordering the transactions should not be the one that is entitled to get a reward for doing so. Therefore, PBS is considered an effective and functional strategy to avoid MEV extraction issues.

VRF Based Sequencing- Verifiable random functions or VRFs sequencing is another security-add on Cero that guarantees fair ordering of transactions once they are considered to be the part of the block. This VRF model is somewhat inspired by the lottery system where an algorithm chooses any random number from the block builder. Based on this number, the algorithm selects transactions one-by-one to be included in the block. The achieved sequence is verifiable by all the nodes before they are included in the blocks.

Sequence Confirmations- The rollup execution process can be lengthy, especially in the case of Zk-rollups, due to the inclusion of additional processes like proof submissions and bridged transactions. Knowing this, Cero provides soft confirmations to rollups, allowing them to perform executions easily or giving users the confidence that on-chain finality will be duly completed without any problem.

Broad configurability options:

Cero is designed with configurability in mind. Be it the settlement layer or the DA layer, Cero enables configuration on all aspects. For example, rollups on Cero can choose a preferred settlement layer across Ethereum, zkSync Era, Polygon PoS, or Arbitrum. This list of settlement layers will be further expanded to support Polkadot, Optimism, Arbitrum Nova, and Avalanche.

Similar to settlement layers, rollups also have the flexibility to choose a DA layer, such as public DA ( like Celestia, NEAR DA, Eigen DA, etc.) or Private validium-based DAC. The first option is suitable for public rollups, whereas the second option is for permissioned rollups.

Try Cero- Get your early access!

Cero is on a mission to offer a highly resilient and future-proof decentralized sequencing network where rollups can operate seamlessly without facing the limitations of either centralized or decentralized sequencing solutions or juggling between dedicated and shared sequencers. To attain its objectives well, Cero is welcoming all kinds of rollups to join the network, experience Cero’s hybrid decentralized sequencing features, and leverage all the benefits we discussed in this article so far. If you are interested in exploring Cero, contact us for early access.

Also, if you have any queries or you want more information about Cero and how it is reinventing the decentralizeds