The last mile of Ethereum validator consolidations

Transcript

[music] Uh okay. Yeah, cool. Uh we have just five minutes here. So let's get straight into the question. Is the last mile of Ethereum validator consolidations is actually doable? And to do that we need a bit of a background. We start with one basic thing about Ethereum validators. And what this thing there are two types of Ethereum validators. Type one the one that most of you know this is a fixed 32e size validator. No more no less. And the other one which is relatively new type two validator where you can start as low as 32 E and you can go as high

as48 E. So in basic terms type two validator can be 60 up to 64 times bigger than type one validator. Great. Why is it needed? Well, because if all of the current stake on Ethereum was controlled by type two validators, we could have had just 21K validators instead of current 800K. And why do we need fewer validators? Well, for simple reason, validators do communicate with each other. And the fewer participants we have for the same amount of stake, the fewer load on P2P network, the faster the network, the more things we can accomplish. So what's the simplest way for us to get rid of this legacy

validators and transition to type two validators? It's definitely Ethereum consolidations. And it is that simple. Imagine you have type one validator. Well, you can always exit exit this validator. You can re-enter with a type two. Boom. Done. But you will have to face exit Q, entry Q. Not comfortable. Now imagine that you also have type two validator. Well, you can consolidate your type one validator into your type two validator. What's going to happen as the result? Well, your type one validator will be exited. So, it will cease to exist. And your type two validator will increase the size without ETH ever leaving consensus layer. Good.

And here are the numbers. So, type two validators were introduced almost two years ago. And as you can see now, everything switched off. So, let's wait for a second. But I can tell you that only 42% as of the end of last month stake on Ethereum network is controlled by type two validators. So type one validators are still dominant. And given all of that, given all that I explained about type two validators, we need to figure out what's the problem. Why people are not consolidating as crazy and why haven't we seen everyone running type two validators? And there's definitely a catch here and it all comes down to the way how automated rewards work. So for type one validator, you

have this fixed 32E balance. All of the rewards that you get above that, forget about this thing being swishing on and off. Uh all of the rewards that you get above 32 E are ever so often transferred to your withdrawal wallet. Comfortable, right? You don't need to do anything. You just operate validator and you get regular payouts. Same goes for type two validator but the limit for type two validator is 24048 ETH which is 64 times bigger and only in this case you can get automated payouts. So you need to choose between 32 ETH like multiples of 32 ETH validators or one big ETH validator and this is really uncomfortable. So is

there a solution? For sure it is. And this is the re the point of this talk. It's EIP8148. This EIP transforms your type two validator from being this fixed monstrosity into a variable thing that is variable when you create it and that is customizable moving forward. So you no longer need to choose between 32 and48. You can choose anything in between. So you can fit the validator to the exact size you need and enjoy automated rewards. So here's a short breakdown of the benefits of this EIP.

First of all, we will finally have one type of withdrawal credentials that fits them all. No need for type zero that we haven't even discussed. No need for type one. It's just type two. Simple and straightforward. Then there's an immediate improvement in the user experience for the new validator because imagine you have 150 ETH. What you going to do now with that? If you want to stake it, well, you will be in trouble. It's not a multiple of 32. So you will either not utilize all of your ETH to solo stake or you will have to go with a protocol and some say that true religious Ethereums should do solo staking. Well, with this EIP you will be able to set the size of your validator to precisely 150 and enjoy automated rewards. There's

also a potential to increase this upper cap of 148 ETH without introducing migrations or whatever. And there is basically no more reason for anyone not to use type 2 validators which now will be a golden standard of validators. So let's consolidate all of the legacy Ethereum validators with the help of EIP8148. Scan the cure code on the screen or go to EIP8148.com and my special appreciation will go to those who will be able to slash their validator at this web page. If you will be able to do it, reach out to me after the call and show me the result. I'll shake your hand. And with that, thank you all.

[applause] >> Dimma, uh, who has any questions? We have one and a half minute for Q&A. is not working. >> Can you hear me now? >> Yeah, I can. >> Okay. Uh, this may be a stupid question, but isn't this worse for decentralization like having a smaller amount of validators?

>> Yeah, that's that's that's definitely a good question and the answer is no because we are not encouraging independent participants to collude. What we are doing is we accepting the fact that there are people who are running multiples of 32 ETH validators like I for example within light of CSM run 22 validators but with this EIP I can easily run just one. So it will not affect decentralization because all of these 22 validators were running on my home node and this new one validator would also be running there. So we are not affecting decentralization. while significantly reducing the load on the

P2P network. >> I think we have a question over there. >> Yeah, we have one more. >> Yeah, over there. >> Yeah. >> Oh, now there's a box. >> Cool. >> Do I just speak to it? Yeah. >> Yeah. Yeah. Speak right into it. So you were talking about the reduction on load for P2P networks right? uh is there any measurable metrics as to how uh what is the like the effect of this when all of the validators move to uh >> yeah for sure uh we are lucky enough that in this case the metric is basically linear uh in the simplest case

and even with the linear assumption we already get an improvement of like 40 times uh or or like yeah compared to the current numbers it's like about 40 uh but honestly provided that like not each node communicate with each other node but nodes do communicate with multiples. So the actual result would be closer to a power function rather than linear. So the decrease will be more than significant. [music]