Accelerating Ethereum Rollup Roadmap by Ahmad Bitar

Transcript

has carries some truth to it. Sometimes that the the Ethereum road map for rollups actually pushes us to uh build and accelerate in that different direction. the properties of a neutral rollup uh technology that is aligned with Ethereum values need some to be to to have these uh properties. First, it needs to be uh permissionless, trustless and censorship resistant. Second, needs to be open source and with an open license so anyone can use it.

Third, it needs to be operable by anyone. Doesn't need to be a data center or a certain entity that operates this technology. Um, and also at the same time needs to be performant and scalable. Introducing search. That's why we came up with Surge, which is a based rollup template that can be used and that showcases Nethermind's client performance. It has four properties. It is stage two from launch. It is based. It's depends on ETH only and it's giga gas. I'll talk about each one and

what it means. Stage two means that um if you deposit any funds into this rollup, there is no way the operators of this rollup will be able to steal those funds. uh uh as long as uh they because they can't really upgrade the the the smart contracts except after a very long period that would allow you to exit if there is a need which is set to be 30 days. Um and this increases the security of the funds meaning the claim about multisakes is not true.

Um it is based that means that sequencing is happening through the Ethereum L1 validators and uh anyone basically can also propose blocks and prove them themselves. There is no centralized sequencers. So it is decentralized. Um, additionally, we only use search uh, sorry, we only use ETH for bonds for um, for for for transacting on the chain.

There is no governance token. There is no VC. It's purely ETH based and we try to contribute to Ethereum by also burning the base fees from this uh from any transactions that happen on this on this uh rollup template. Also, it's gigas meaning you can reach a throughput of 1 gigas per second or more with the performance of Nethermine client without any issues. Um so how did we reach these these these um these properties? We had to make certain design decisions. First we use

the Tao stack. It's open- source MIT license based rollup stack. So we choose we chose the tao stack and then we took it and we replaced the finality gadget instead of what tao has we changed the finality gadget to depend on a two out of three provers. We have sgx sp1 and risk zero. Uh risk zero and sp1 are zkvms zk provers and sgx is a te. Now to prove a block what you need to do is to provide two out of three of these proofs for

this block to be proven and verified. Once it's verified you can basically bridge out in the case that there is any um settlement that is needed for bridging. We also replaced the token type so and the the that is needed for the bond. So for Tao you need to uh supply Tao tokens for the bond. So we remove that and instead you uh you you put uh some ETH for the bond uh when you are proposing um we also changed the split of the fee.

So uh we are burning 100% of the fee whereas in Tao it's uh a lot of it is going uh to the proposer and a little bit of it goes to the taod dowo um of course we ch uh we we replace the execution client and the execution client the nethermine client we we we the execution client for tao stack They use normally Tao Gath. Um but we replace Tao Gath with using the Nethermine client uh with the Tao integration. And then the last step was

to add extensive documentation for how to operate the chain, how to launch your own devet and roll up on top of it and how you can even uh operate uh the the the live test net if there is one. So what can you do with search today? Well, you can go to the website and check out what we're doing with SER. We can you can also um uh see what we're planning to do and the future roadmap.

You can operate any component of the search test net that is live right now um by yourself uh on your on at home or at on any machine. It's completely permissionless and you can deploy as I said your own rollup uh if you wanted to you can just go to the docs p uh to the docs uh website and you will find guides on how to accomplish that extensive guides.

Um last you can check out the code which is available on GitHub fully open source. All of the repositories are available. Um and last we you also can go to the bridge and bridge some testn net e and try things out on search transact deploy some daps play around with gig gas see how it goes. Um so what's coming up for surge? Um so what we're planning to do is to work on the pre-confirmations. Uh because right now

the UX for surge is that you um you have to wait around 12 13 14 seconds up to 20 seconds for your transaction to land because it depends on net landing on the L1 and that's bad UX right so what you need is to have preconfirmations which could potentially provide you with 1 second or even subsecond confirmation times we're also working on composer ability where you can in the same L1 slot pro uh bridge bridge to the L2 um or to the rollup transact on the rollup with these brided asset and even

bridge out in the same exact slot and we're expecting to get that in a couple of months. Um we also are thinking about a better fee structure where um the L2 fees actually uh represent the actual L1 cost for proposing blocks on the L1 and you can get a better UX based on that. And last, we are also looking into prover and finality gadget improvements that could increase the security of the user's assets.

Um, we're also planning to deploy Serge on mainet and we're hoping that the community will pick Serge up and we'll take it over and operate it by themselves without intervention or any help from Nethermind but like no govern maybe engineering help from Nethermind but governance and everything else will be fully supported by the community. Um, so this is a little bit technical. We're going to move into the finality gadget. So what we have right now is that we have every block needs

uh needs two out of three provers as I said. Uh the improved finality gadget as displayed here says you need at least one TE prover which is SGX or TDX which we are working on developing with Nethermine and one ZK prover zero or um uh SP1 and if you can't get so this new finality gadget depends says you you need to have both so TE one of them and ZK either one of them. And if you can't get both, what you can do, for example, because of a uh

uh prover not being able to prove a block, what you can do is you can do an OP proof. So you submit a single proof for your block and it it it starts a cycle where it waits for 7 days to see if there is any contradicting proof to this one which would eventually then be finalized or someone would actually contradict this proof and you would end up in most cases also being finalized.

But there is some cases where you need to wait for two proofs TE and ZK that are matching to be able to finalize a block. And these are when there is um a fault in one proof system where the proof is contradicted by the same proof system later on. And that means that there's something wrong and we need to wait for a third proving system that can help adjudicate. We still don't want any security council or multisc to intervene

in the finality gadget. Their role is to only be able to upgrade in the case that something goes wrong with one of the provers and the upgrade needs certain conditions to be an onchain provable bug otherwise this upgrade is not allowed. So what does uh what does Ethereum uh what does uh community takeover of surge once it's deployed on mainet looks like? There is an initiative by the community. It's called Ethereum R1. It's basically a new rollup uh

project uh trying to align with Ethereum rollup uh neutral infrastructure. Um it it is fully donation funded. There will be no VC funding and or private funding and um only ETH contributions. It will be also an Ethereum public good where the DAO is control of all of these funds and it it supplies the engineering team of Ethereum R1 with funds to continue development and it's replaceable by design. So there should be no

um proprietary governance or um uh it it should be like anyone should be able to take over uh in the case that there is a need for for taking over by the uh uh by the community. Um yeah and this this is not announced yet uh officially. This is uh coming up soon probably by the beginning of May. There will be a donation um there will be a website and there will be like a link for donation for the community. Whoever donates becomes part of the DAO that basically governs this um um is is

participates in governance of of this rollup. And that's all from me. Um if you have any questions, I'm happy to answer them.