An introduction to sequencing
Last updated
Last updated
An Ethereum validator secures Ethereum. It performs several duties but for learning about based rollups only the block proposal duty will be discussed. Block proposals are random and rare, on average once every ~6 months but they provide the majority of the rewards a validator receives.
Even though Based Sequencing has only more recently been discussed in the context of L2s, the Ethereum L1 has always been a sequencer.
When a validator is selected to propose a block they have complete control over the order of the transactions in the block they build. They can't include invalid transactions (tx) but can include (or exclude) any tx as long as it covers the base gas fee.
This may change in the future with features such as FOCIL, BRAID, and Execution Tickets but as of writing, Ethereum validators still have absolute control during their block proposal.
When a validator only uses their local mempool to look for txs to include in their block it is called vanilla block building.
And that's it. A validator can be a full participant in the Ethereum network just by performing its required duties (attestations, etc.) and proposing vanilla blocks. So why are only ~7.5% of blocks vanilla?
Maximal extractable value (MEV) is a huge topic but from the perspective of an Ethereum validator, it's simple:
Someone will pay for control over the exact order of transactions within blocks.
Why does the specific ordering matter? Ethereum already has a way to get transactions included in a priority order. The priority fee can be set higher so that validators have an incentive to include a particular tx first in a block. But there are cases where even greater control over the tx ordering can yield much higher rewards. This could be simple arbitrage between a centralized exchange and a DEX or a more malicious sandwich attack, but if there's money to be made (and there's a lot of money to be made!) then someone will pay for it.
This is where MEV Boost comes in. Finding the optimal tx order to maximize MEV is a highly sophisticated and resource-intensive activity. To allow any validator, even solo home stakers to participate in this highly competitive endeavor, a block-building marketplace was created. All a validator has to do is run the MEV Boost client and tell it which relays it wants to use. From the perspective of a validator, this is a win-win. They can opt-in and opt-out at any time, it's free to use, and they get paid significantly more for each block proposal! There's even an option to set a min-bid parameter so that if a validator receives a low offer for your proposal from the MEV Boost builders, they can still build their own blocks locally.
Free to use
Opt-in and opt-out at any time
Set min-bid value as a fallback for local block building
Centralizing effect due to hyper-specialized block builders
When a validator uses MEV Boost it must blind sign the block header without seeing what will eventually be included in the block. This can be risky. If the block builder submits an invalid block or doesn't even send the full block, the validator will miss their proposal. Relays are used as the mediators for multiple block builders, and can check that blocks are valid and give validators a trusted point to query the current highest bid for their proposal slot.
Notice that multiple block builders can talk to multiple relays and that a single MEV Boost client being run by a validator can select which relays it connects to. There is an auction for each block with block builders submitting their highest bid and the relays sharing those bids with the validator.
Ethereum is a sequencer for its own transactions. A validator can choose to order the txs in their block proposals themselves or outsource it to specialized block builders through MEV Boost to significantly increase their rewards.
But what if the sequencing concept is extended further? What if instead of sequencing only the Ethereum L1 txs, Ethereum validators also sequenced L2 txs? This is Based Sequencing and is the foundation of Based Rollups.
That looks like a lot of extra work for an individual validator - but don't despair! In practice, this specialized L2 sequencing, proving, and blob creation is performed by specialized actors and the proposer simply accepts whole Ethereum blocks through MEV Boost. If you are an Ethereum validator currently using MEV Boost you are already sequencing based rollups without even knowing it 🤯
This is how Ethereum scales. This is how Ethereum solves fragmentation and composability. This is how Ethereum becomes the base layer for everything.
One of the main benefits to users currently on L2s (other than much lower gas fees) is that transactions get confirmed very quickly, on the order of 2 seconds or less. This is a big UX improvement as users get a ✅ response without waiting a full 12 seconds or more for their transaction to be included on the L1.
So what is the solution? Preconfirmations ⏱️
Preconfirmations will require a validator to opt-in to say "I'm the next block proposer (or soon will be one) and I see your tx in the mempool. For a fee, I will promise to include it in my block, so you can get a ✅ immediately". Before explaining preconfirmations in more detail, let's first look at how rollups with non-based sequencer work and why based rollups would want to emulate the best parts of that user experience:
Can decrease censorship resistance of Ethereum ( helps with this)
But there's a catch. The Ethereum L1 is slow!
