RFC: Proposing a New Mechanism to Decentralize Sequencers - Distributed Sequencer Technology


Greetings Obol Community,

I’ve been researching DVT for a while now and recently I have been considering ways in which DVT can be applied beyond Ethereum’s consensus layer. A few months back, I had the realization that DVT might be able to help solve the ever-growing problem of sequencer decentralization.

As many of you know, rollups today are centralized and fail to ensure appropriate censorship resistance and liveness guarantees. Despite the steady demands from the community, rollups have yet to decentralize. Many have not even signaled their intention to. Sequencer decentralization is a complex matter that will take time to solve. A number of rollups will not choose to decentralize even when solutions are available. Solutions which can progressively decentralize rollups today while also allowing rollups who wish to remain centralized to do so present a high degree of viability. I believe DVT can be applied in the form of DST to accomplish exactly this.

Proposing a New Mechanism to Decentralize Sequencers - Distributed Sequencer Technology.

Today, validating on Ethereum isn’t simple and comes with considerable risks. Distributed Validator Technology (DVT) removes the single point of failure risk Ethereum validators currently face by distributing the responsibilities across a cluster of node operators.

DVT helps further decentralize Ethereum’s consensus layer while also promoting greater client and geographic diversity amongst the Ethereum validator set. It improves liveness guarantees and offers a fault tolerant solution for node operators.

As we dove deeper into DVT, we wondered whether the same methodology could help decentralize the current centralized sequencing process and lead us to investigate whether Distributed Sequencer Technology could help progressively decentralize L2s.

Sequencer decentralization won’t follow a “one-size-fits-all” model. Single sequencers are problematic for most rollups considering the single point of failure risk. For rollups less interested in decentralization, it still offers benefits.

DST’s ability to achieve fault tolerance while also promoting greater client and geographic diversity provides an important improvement in liveness. It ensures rollups continue to operate even when one or more components fail. This is an almost universal need across rollups.

DST provides a step towards decentralization; rollups can share sequencing duties with more third party operators. The setup shares resemblance to PoA and helps reduces the risk of censorship for a given rollup.

DST does not require rollups to decentralize. In a distributed sequencer, a central organization can remain in control of sequencing by operating the entire cluster. These entities can forego the PoA-type setup described earlier and remain centralized.

Through DST’s flexibility, rollups can approach decentralization as they choose and still achieve the liveness guarantees they need. A game studio’s app-specific rollup might remain centralized and a general-purpose rollup might use it as a step towards maximal decentralization.

In either scenario, DST shows significant promise in helping to solve several issues that have been identified in sequencing today. We believe DST is a practical solution that L2s/rollups of all kinds should consider implementing.

For most Ethereum-equivalent rollups, like Optimism, implementing DST shouldn’t require a heavy lift. The rollup has already adopted most of Ethereum’s infrastructure and code. BLS signature support is one of the few changes needed to implement DST on the L2.

DST offers L2s a promising new way to incrementally decentralize. Regardless of the features a rollup is optimizing for, the benefits from adopting DST span across several preferences. We are excited to introduce the concept of DST in full detail here: Distributed Sequencer Technology — A Path Towards Decentralized Sequencing | by Figment Capital | May, 2023 | Medium