CSM v2: Empowering Home Stakers | Dmitry Chernukhin | ETHWarsaw [4]

Hello everyone. Um, my name is Dimmitri and I'm really glad to be here today. Um, I'm I've been contributing to Widow for a few years now and today I want to share with you something that was a massive part of this time. But um, this is um, the community technique model or CSM as we call it. But um, that's the thing is it's not what it exists today.

But I was talking I will talk about what is uh coming next. Uh so for those of you who might not know CSM we um CSM was launched in October 2024 with simple goal in mind to make Ethereum staking more accessible. So instead of needing of 32 if you just need to bring 1.5 if to create your validator and the best part of it is that you're not sacrificing on rewards and you actually uh um was getting probably the highest rewards on the market. So uh CSMV1 was a massive success and our pro our team is really proud of it.

Um the question is why CSMv2 and the answer is that um it was known from very start that the single model cannot serve for everyone and actually we need to develop a more granular system that will cater to different types of stakers. So uh the goal is to create a truly sustainable home staking model and to recognize and reward um contributions of individual stakers. So I'm really proud to announce that the expected goal live date of CSMV2 is October 3rd of 2025 and it definitely worth the wait because of the features we just will be talking about today. Um so first of all what does this more granol system actually means? Since CSM v2 there are three uh officially recognized type of stakers.

First of all identified community stakers. These are the backbone of decentralization. These are the homestakers community stakers. Then uh legacy early adopters. These are the operators that join CSM v1 during the early adopter early adoption period.

Uh these operators have proven their commitment and they will retain their special status and will have a beneficial bond requirement similar to the identified community stakers. And uh finally the permissionless operators um they uh have been the default category in CSMv1 and basically everyone can join the CSM without any special requirements and in this way the system ensures that it remains open and accessible to everyone and um the system was designed in the way that it will support professional operators even though they're not a part of this SMB2 launch. uh because the main focus remain on the identified community stakers. Um the system is flexible enough to accommodate professional operators as well and that's why you will be able to hear more about professional operator solutions later this year at defcon. So um how to actually provide the differentiated treatment to different operator types and the answer is the our parameters register system.

You can think of it as a control panel with different parameters that you can set for each operator types. So this is about different rules, thresholds, rewards etc. And this means that the experience can be customized based on operator type and what it actually brings to the network. Um so let me walk you through the exact uh parameters we were introducing in CSMV2. First of all uh this is a performance related parameters uh such as uh thresholds and leways.

Then economic ones uh such as reward shares and penalties. And then operational uh this is related to uh priority cues and strikes. We'll talk about it like momentarily. And finally security parameters. This is about bond requirements and the exections thresholds.

So first of all about the probably the most important one parameter this is about performance coefficients. So in SMV1 uh a performance of a validator was just um a relation of the attestations assigned to attestations included to attestations assigned. So this is pretty straightforward. But in CMV 2 we now um account for all of three duties um of a validator. So artistations, blog proposals and scene committee participations.

All of this is all of them accounted to or determine the performance and all of these duty have a have a defined weight and by default um it reflects the consensus layer reward structure. So attestations carries the most weight followed by block proposals and finally sync committees and it can be customized per operator types and it means that identifying community stakers can have more linear coefficients uh for for example for blog proposals and sync committee participations to um better reflect the reality of home staker setups. Okay, building on these performance uh coefficients, let's talk about performance leway. Uh performance leway is uh measured relative to the network average performance. So identified community stakers will have a 5% buffer.

So they can go 5% lower than network average performance and still get the full rewards. um it is applicable for the first 150 validators but I like it's reasonable to assume that uh community stakers will fit into this threshold anyway. So and for permissions operators it's just a 3% buffer and it is it accounts for the fact that professional operators um I mean otherwise community stakers operators don't face the same issues um as professional operators do. So for example, internet orages, equipment restarts um and other connectivity issues. Okay, let's move on to the economic uh parameters.

Um that's where is the differentiated treatment um comes in again. So identified community sters will get up to 6% of the total allocated sting rewards for their first 16 validators. So it creates that direct incentive to be identified as a community stakers and with and the rest that permissionless operators will get uh flat reward share for across all of their validators. Okay. Um with all of these benefits we actually need for some kind of uh quality control mechanism and that's why CSMB2 introduces the automated strike system uh to maintain network qual quality while being fair to operators.

So our performance oracle will constantly moni monitor um the validators performance and assign strikes to them whenever they fall the acceptable thresholds. The question is was actually what is actually trig what actually triggers the strike and the answer is the persistent poor performance. The key word here is persistent meaning that the small hiccups won't u won't be penalized and instead system look for patterns of consistent underperformance and that's exactly why strikes have lifetime. So strikes will accumulate and eventually they will expire after or several oracle frames on mainet is supposed to be 6 months and it means that like past issues don't haunt you forever. Um I I have my personal experience with it.

So I'm running a validator in CSMV1 and u I was traveling and basically on day two I was realized that my validator stopped working. I was trying to connect to it remotely and basically nothing. So I have no idea what is going on and it's in Cesm actually I will get a strike and it will record it but the best part of it that after 6 months it will expire and this bad week won't stick for me forever and I will be able to um correct the issue. Okay. Um so this is about thresh strikes threshold.

So this parameter defines how many strikes a validator may accumulate before it's being ejected from the protocol. Uh so the system works automatically and when your validator like reaches the threshold the permissions method on the smart conduct is being triggered and your validator is being ejected. Well, that's easy and the best part of it as for the most of the parameters that identified community takers will get more leeway. So we they will be able to take more strikes and still um keep their validator in the protocol. Okay, now let's talk about how to actually get into the system.

So there are two different ways and entry gates and extensions and they work quite different. Uh first of all entry gates uh they work they are quite straightforward. They works as a simple ID checker. So they verify your credentials assign you specific uh operator types let's say identified taker and then you can interact with CSM directly without any other smart contract. So you just claim your um specific operator type and that's it.

So it is pretty straightforward. Extensions in turn are full proxy system. So um I was like this uh duckness because uh my analogy here this was like you calling uh your broker um and saying I want to buy a pile of stocks or something like this and they do the job for you. So the extensions in fact um the all the interaction with uh CSM actually goes to the extension itself and in this way an extension can implement some kind of internal logic for example they can implement native uh DVT cluster support or specialized reward sharing etc. So this is much more complicated uh compared to the gates.

Okay. Uh let's now talk a little bit about emergency situations. CSMV2 uh introduces support of EAP72. This is about execution layer triggerable exits. So uh and this is a crucial recovery mechanism for uh homestakers.

Whenever you get lost access to your validator, let's say you have a corrupted private keys uh or you just logged out of your setup, you still will be able to exit your validator by interacting with the CSM contract directly. Uh and UI will just expose a simple emergency recovery button that you will be able to ch to use. Even though it comes with a fee, it's still a crucial um safety net for your validator. So you it means you won't be permanently stuck with running validator you can't control and um the same EP72 uh is used by the protocol for their own security means. So first of all our validators can be triggered um for exit if they underperform for quite some time.

So as I described before uh when a validator accumulates enough strikes so it has to be ejected from the system and the second one when there are validators that have to be exited uh because of withdrawal demand to meet protocol obligations when the operators like can't or don't want to act uh like uh by themselves. of the protocol and this way uh like can control validators that inside of the module and um yeah let's talk about priority cues. This is the thing where the differential treatments really shows because um identified community stakers will be able to get your validators their validators activated faster and that's how it works. So every operator types will get um predefined um priority level and um when the protocol will need to uh deposit if to create new validators, it will uh process the queue in order starting from the highest priority first. And in this case um identified community stakers will be able to get um validator slots basically before the others like permissionless ones.

Uh to put it into perspective like in CSMV1 when it went uh fully permissionless um big node operators just fluided in the system. So there were operators uh uploading like 200 keys and um small home stakers were squeezed out of it and I was watched like someone just stuck behind um a large operator with 200 keys uploaded and CSMV1 actually CSMV2 actually changes this. So now identify community sters will have will have uh 10 priority uh slots that will be deposited before all this large u permissionless professional operators and this is I guess one of the most important um things that uh brings um to to stakers. Okay, let's talk uh about like the all benefits that uh ICS brings. First of all, um atified community stakers will get increased rewards that design it uh to um to cover hardware cost within a year.

So a staker with 16 validators can get like this increase the worth as high as 0.4 if per year. So I guess it kind of enough for covering I don't know electricity bills, internet bills etc. Um and of course it is still probably the lowest bond requirements on the market. So to create your first validator you just need 1.

5 E and for the consequent ones it will take only 1.3 and of course the priority Q that I was just describing. Um so it means that community stakers will get deposit first at least for their first 10 validators but I guess it it's it's really good one. Okay. Now you may be wondering how do you actually get identified as a community staker and this is about identified community staker um framework and the best part of it is that it is secure transparent and it does not require like some kind of KYC if you don't want it you you can omit it we do understand that privacy is like paramount and the system is designed in a way that it uh respects it uh the framework is built on three core pillars Uh first of all it is a proof of engagement.

The system checks your participations in governance uh open source projects etc. The goal here is to identify active members of communities that really uh that are really committed to the success of Ethereum and LAM. Then uh proof of experience. So that's about um have you been running a validator before? Have you have you been participating in test nets etc.

So your experience in staking and finally and probably the most important one is it's a proof of humanity. So the goal here is to prove that you are actually a unique living person and that you're not trying to create a some kind of a hive to um to just implement some kind of the civil attack or something like this. Um so yeah this three call appears and uh the IC framework is live so you can go to I believe to the CSM web page and send your applications to be identified as a community staker. So let's let's wrap it up. Um we really proud with our job and we think with that CSMV2 will help to make Ethereum even more secure and decentralized.

So as a team we believe that it will help to significantly increase the number of independent operators with this new beneficial uh terms for um stakers. We think that over 10% of lighter will be eventually controlled by community stakers. Yes, this is kind of bold statement but still we believe that it's achievable and that the total number of light node operators will surpass 1,000 and it will even more diversify the node operator set and lighter and finally that Ethereum staking will become a common hobby for everyone and the last actually last one slide probably that um we're not stopping on it and we are currently thinking about CSM and future features that we can bring in there. And we are thinking about compounding validators and support of uh increased max effective balance. It will probably significantly increase uh rewards for uh larger operators, larger validators.

Um yeah, so we're also thinking about enhanced professional operators features. So as I said more about it later this year and maybe on the data from the CSMV2 we'll decide to add more general operator types to like better solve the system and that's it. Thank you.

Do we have any questions this time?

Um hi thanks great co uh talk. Um you mentioned that um the strike based ejection is permissionless. Is this also like it's also the the oracle data required for that permissionless like some proof on the on this on the state I guess.

Yeah sure. So um the Oracle will bring just a tree with leaves that says this node operator this validator uh has these strikes and uh you just have to generate a proof that this leaf was included in the tree and uh the dedicated conductor test will accept it and if it accept it it triggers the objections. So so you're right that's exactly how it works.

Thanks.

Sure. Do we have more questions?

Good.

Cool.

Perfect. Thank you very much. It was a pleasure hosting you here.