# Ethereum Staking from an Institutional Staker`s View Florian | Lueffe (March 2024) - Channel: [Berlin Ethereum Meetup](https://streameth.org/berlin-ethereum-meetup) - Date: 2024-10-07 - Duration: 31:50 - Watch: https://streameth.org/watch/yt-SXpNATy2J5s - YouTube: https://www.youtube.com/watch?v=SXpNATy2J5s ## Description Join us on Meetup to keep track of our events in Berlin: https://www.meetup.com/berlin-ethereum-meetup/ See you at the next one! --- Apply to speak at our future meetups: https://forms.gle/5y9Y5ywZC7pSEqpV9 --- Twitter: @BerlinMeetup ## Transcript all right so let's take a little bit about ethereum staking from our perspective as an professional Staker um yeah I will give you a short introduction about me about blockscape uh and then we we will jump straight into our staking setups at blockscape uh I will also have some best practices for you if you are a solo Staker and also give you some advice if you're not yet in Solo ster to become one and in the end I would just talk a little bit about the future of E staking uh specifically I want to talk about two cool Concepts one is reaking and one is Rainbow staking and then we will have our conclusion and you can ask me questions if you want all righty so let's jump in GM I'm Flo and I'm working as a blockchain architect at blockscape uh that basically means I have two tasks the first one is that I build stuff which means I break our infrastructure and the second one is I support our devops team and fix the stuff I've broken before for basically okay and who's blockscapes so we are a small team based in Germany and Switzerland um our main businesses being a validator in blockchain ecosystems uh we are active in ethereum and in like 15 other Cosmos chains uh we in eum we're are part of Lio and SW liquid staking operators Set uh we are also doing DVT with obol and SSV and I think we have currently 10,000 uh active validators but we are actually doing a little bit more than just staking operations so we have our own liquid staking platform called unake it's based on Stak wise V fre setup so it's a pretty cool uh concept actually because everyone in this room could B basically spin up their own uh liquid staking token with steak wise solution uh we also built a UI for Gravity bridge that is a bridge between between Cosmos and uum uh we are active in the community and also we are small team we are doing a lot of research and development and always looking at the next Innovations okay so let's jump into ethereum staking so I mean we all know probably that staking is a core function functionality in a proof of stake blockchain seems reasonable and maybe we are all aware of the blockchain trilemma uh basically that means that's pretty hard to achieve a system that's scalable secure and decentralized all in the same place so you could always reach security scalability but then decentralization is very hard to get there and all the other Corners like that so um ethereum is doing this pretty well in our opinion but there are some centralization RIS so one is the LST and sex clumping basically and as a professional Staker we are kind of part of this um there is that M Builder centralization and we also have that client centralization but luckily there as Max already mentioned uh it's getting a little bit better the cons consensus side looking already pretty okayish and the execution client yeah I mean G is one big issue here but I think that's something we all can actively work on and we as an operator for example do this we move mainly moved mainly to nevermind and basil and are still moving our G setups to to other execution clients there's also Reef that's a new execution client written in Russ made by Paradigm that's pretty cool and yeah that's something we can act actively work on and yeah regarding this LST and sex Plumping I will actually come to that uh in the end with rainbow staging but might be a cool concept against this okay so then let's jump in our staking setups so from an basic P perspective how is a basic or a general staking setup look like we have our execution clients that could be gav NE Basu Reef or whatever we have our um Beacon client which could be prism Lighthouse and if we want to run a valy data we also need a valy data client which could uh it basically often comes U delivered or packed with um the beacon client but theoretically speaking you could also mix and match between a different uh valid dat client and a different Beacon client um this is also the place where your validat a key would be normally stored and and then nowadays you maybe would like to run me boost that means you will be connected to me relays and one simple Improvement you could add to the setup is a web free signer a web free signer is basically like so it's comes with a database with a postgress database and it basically moves that key from here to there and it makes the key management a little bit more easy there will be a yeah there will be a connection between the web freeer and API connection and the valid data client and it generally is a little bit better you could also move that web free signer outside of your main validating setup so that it's for example an offline system and only connected to your main staking setup that would also be possible but at blockscape we are doing things a little bit different um so most of professional stakers do this but we using threshold signature schemes for our um signing so that means uh again we have our basic setup and we have that free signer but now we split basically the setup in uh other clients that are called duk and each Duke client is using so they're using um the cool thing in the beacon chain is you have that BL BLS schemes and they support Shamir secret sharing so we can just split your key in a similar fashion how DVT works and um then you also need a validat client replacement the solution is called vouch and all the rest is basically the same and dur and vouch is a software Solution that's made by attest and it's pretty cool because this allow it allows threshold signing so you do not need to store your P valid dat key in one place instead you just have the key shares on different um uh Duke instances in this case you can also choose different threshold schemes so you can do three out of uh two out of three uh three out of five Etc and this would be probably that's a typical production setup for us so we have like uh fre servers uh in three different locations uh or data centers and then you have on each of the first two you have a consensus and execution client and the Duke instances and the last one is only a Duke and the vouch so one disadvantage in this setup is that you can only run one vouch because vouch is doing the stations all the valid data duties and you cannot run two or more because that would lead to a double signing and in worst case to a slashing then um that means if the first server goes down for whatever reasons we actually do an fail over to another instance so we will actually delete the server in the first place then do the fail over and just spin up a vouch instance on one of the other then we have time to fix this but in the meantime the validator will still do its duties so that's a pretty neat setup uh yeah okay so what are the advant advantages of d d and vouch and the threshold signing scheme so we have an increased security because there's no single point of fature we have increased re silence one note can go down and we are still doing our duties we have at least the technical decentralization here and we have that for tolerance uh the tradeoffs um it actually adds some complexity to the key management now we do not have just one key for one validator instead we have three or more depending on your threshold you're choosing and it could have a negative put um performance impact so higher threshold uh adds a little bit of latency to the setup which could be worse in your Effectiveness rate and you need uh secure communication channels between your notes so it is not something you could actually do with third party operators or with other guys so you need to have a private net where the notes can talk to each other and therefore it might be not the best solution for home staking okay so but now there is that awesome technology called distributed valid data technology and Max already did Tau you amazingly how it actually works so I will just skip this and give you just a short comparison between SSV which is one of the service providers or yeah developers of DVT solution and obol uh yeah and you know obol already so and also Max already Tau you um the SSV node as you can see here actually replaces the value data client um the SV note uh will handle all the charts uh the the the key shares of of your SSV setup or DVT cluster and will do the peer-to-peer connection to the other cluster members and will do the valid dat duties the rest is looking the same um in comparison or bow and that's pretty cool in my opinion uh the Corone client basically just plugs into the valid dat and Beacon client and you can still mix and match and have awesome client diversity because you could run for example an execution layer G with nimos and then Lighthouse as in valid data client which is pretty cool in terms of yeah res silence and client diversity um how we do you how how we use um SSV uh specifically in our case so as I told you in the beginning we have our own liquid staking solution called unake and um we actually build it on top of DVT so we spin up four notes that is one DVT cluster one is in Canada one in France one in Germany one in Singapore and they are talking to each other because they will do the validator duties for our uh onboarded validators and performance- wise it's crazy how good it actually works because I mean it's completely geopolitically decentralized and still doing awesome Effectiveness rates you can check it yourself on rated and yeah it's cool and another another way we use it is um we spin up clusters with other node operators so that you do not just have that geopolitical uh decentralization but actual decentralization in terms with the entities you stake with so here you can see that's a cluster of 400 validators and um the Staker he or she um just choose this random validators in this in this set and we are doing the validator duties with this random or not random but with this note operators here and that's also pretty cool because now you do not need to trust one entity you need to we all need to be Reliant and at least three of them need to do their Duty that the validator is performing well but I think that is also pretty cool okay so then we will go to the best practices for solo staking so um I will just go into detail for two main uh Solutions because I think they're are pretty cool and differentiate from each other the first one is dep note you probably or you maybe had heard of it it's a very user friendly plug andplay solution uh you can buy their Hardware completely installed or it's completely open source you can also install it on an old laptop or notebook or whatever you have that requires the hardware uh requirements for an etherium Staker uh it has a cool Telegram and email integration that means if you validated goes down or there are important updates you can get an notification very easily it has an web management web management UI so everything is uh graphic with everything is made with a graphical user interface and you also have a sort of App Store so they they call it dep store and you can simply select your prefer PR preferred staking solution so you can go with ethereum staking you can do gnosis staking I think you can also run a Bitcoin note yeah that's pretty cool and on the other hand uh I would want to talk about a little bit about anible anible is a lot more advanced and en is not something made specifically for staking purposes so it's a general open source framework written written in Python it's it's basically for the automation of um yeah um setups of operating systems and uh you can basically do anything with it it have a it has is it has a steady learning curve um you can run anything and one very very cool thing is that the e panda Ops have made an Enz collection and this makes us insanely easy to spin up your valy data with it so what I will come to this in a second and there are also a lot of other Solutions so this are just two examples and here are just yeah other ones you could use so they are all pretty cool um just two screenshots for dep note so this is basically how it looks like this is from one of our colleagues he's um running in dep note at home home uh here you can see he's basically doing osis chain validation uh you can see how the client is working and here's also the dep store where you can for example also go with rocket pool staking you have you can click on get and then you have a cool installation visard so it's extremely simple um okay enable enable as I told you is pretty cool because this 30 lines of code will actually spin up a completely working ethereum validator for you that's you need to do you can even choose which clients you want to use and it's yeah it's it's extremely simple you just need to add your fee recipient you can put your valid data key on it and Bam you have your ethereum ster at home and one thing uh because so one personal thing I also used to use um de node at home but I had one problem one major problem I wanted to run an SSV node as home at as well and um the disadvantage of the depot ecosystem is if there are new solutions they need to firstly adapt it to the ecosystem so that they run with a nice UI and all that stuff and yeah I wasn't willing to wait for that so I switched my home staging setup to anible and the cool thing in anible as I told you is you can just use if you want to run SSV with it a DVT DVT solution the line 14 is the only only thing you need to do uh you need to configure the SSV Keys okay but that's all described in our repo so you our this repo is open source you could use it if you want to do it on your own and with this you have an DVT home EV DVT note at home in like yeah seconds and that's pretty cool in my opinion and yeah one one approach you could follow with this is to set up an multis like safe with your friends uh create uh especially for SSV create your valid data key split it um create the SSV cluster with your friends and one cool thing in my opinion also is that you can add an Institutional professional Staker to your setup so that would basically look like that uh that's the SSV UI and you can see here that that's my home SSV Staker that's just another random uh home Staker probably and then I just choose two um institutional professional stakers that will bring better fault tolerance and better res silent to my home stacking setup which is pretty cool in my opinion and you could also do it with different cluster sizes like four 7 10 and three and depending on the size you have different four Toler rances so here in four it's one 2 three and I think four then yeah probably um okay some tips for monitoring and earting at home so I would always suggest to keep it simple but functional uh as I mentioned you can just if you're going with depote you can just use their telegram email notification that's awesome because you know your valid dator is doing at stations you get a block proposal and you are happy or you get a notification that your Val valid data is down and you're not happy proberly um you can set up DMS that is uh the monitoring system of dep node it's also Plug and Play you can just install it and everything works and in anable you can use grafana and prome for promos that's the monitoring service basically there's also an anible Playbook made by the e panda op so it's also very Plug and Play is and one thing that's also pretty cool is the beacon Chen app it's free to use I think it's free to use for like 100 valy datas but I think if you have more than 100 validators you could buy the app probably um and you get um with that app you get basically notifications of the monitored or of the um of selected validators you want to monitor and that's also pretty cool but because you most likely have your mobile phone with you all the time and then you know then you know what is going on with your validator um and one very exciting networking tip so I had this problem on my own so it's very simple basically to fix but very uh yeah very um very uh unfortunate if you do not know how to fix it um if in your home rooting device um You probably have that option to enable universal plug and play or in short UPnP protocol that basically will um help the client software of your staking server to automatically fetch the port forwarding to the outside world so that you have the connection to the other peer-to-peer nodes and so on and that will uh improve your validator performance and your um no peer Discovery extremely and the second part is this a little bit depending on your internet provider but uh we recently moved in uh Berlin and our new internet provider did not assign us an public IP V4 address so the only thing I need to do and it was free of charge as well I need to call um him or her and told them yeah I want to have an ipv4 address uh 5 minutes later I I was assigned a new address and Bam my peer Discovery went all the way up so that's also a very important thing uh you can check it at home and yeah if if you want to if you want to run a home Staker that's probably a very good um starting point or you can look this up and fix it probably at your home all righty so finally we will jump to the future of ethereum staking um and as I told I have two topics the first one is reaking so reaking is a pretty cool concept um the eli5 would be that you can um reuse your sted if um in other services outside of ethereum so you have your ethereum blockchain here and here are other services so it could be everything it could be Oracle Services it could be data availability yeah probably anything you can think of uh these servic are called actively validated services and yeah as I mentioned they live outside of the ethereum blockchain most likely and from a workflow perspective so imagine you have your LST like rocket pool e or Lio s h you can just restake it into the smart contract so that means you will just deposit in the I layer smart contract and from there on you can delegate it to different IG layer operators and then the operator and that's also one disadvantages in my opinion the operator can select the avss he or she wants to uh or he or she wants to secure so you cannot directly choose the avss you want to support and want to um give the security to you can only choose your operator you you can UND delegate and delegate to another one if you're not happy with the choices but that's the current way how it works uh and not only as an lstd Staker you can go live or restake your assets in i l but you can also do it as an ethereum validator um that would mean you need to spin up a new validator so you cannot use your existing one once at least if you already have set your withdraw credit credentials in it because you need to create an iport address with the layer smart contract um and then put your withdraw credential to that iport address that will basically restate your 32 e and your valid dator and then you can also delegate it to the operators and then there's also the new cool thing around the corner called LRT liquid reaking token protocols like EA EFI H and they do basically the same thing but instead that this is in validator they have a smart contract and the smart contract creates the eyp and then restates the um your e or your deposited e in the smart contract of the L LT protocol and then it will also delegate to one of the operators and here you probably have uh only the choice to um delegate to a specific operator via an Dow Vote or something like that so depending on the protocol you are choosing yeah but um in general reaking could be very interesting uh it would it will bring more risk to the staked assets but it potentially will also bring more rewards to the assets uh to to your Eve because you actually get yeah the rewards from the services that are doing stuff here um so could be pretty interesting in the future and we are uh looking into it as well so for example if you want to just uh play around with it you can go to heski do .xyz uh and and just uh deposit your Stak or your wrapped eer in heski for example and yeah have a look and feel how it will go on mainnet in the it will probably go live in the next month uh uh yeah okay so and then uh I just will slightly go over the rainbow staking proposal and if you want to have a deep dive into it I can highly recommend this QR code with the E research post um but in general rainbow St is a new proposal made made in the Forum um and the idea is to widen the spectrum of um possible participants in the staking uh ecosystem so nowadays you have that one validator service and everyone needs to run just one validator service so there's no choice in Services you can or in duties you can actually fulfill as a validator and with this idea you can choose between different Services you can run know different duties you can run as a Dy data there are heavy Services um which could be the consens consensus mechanism maybe in the future things like finality the block production in general uh it could also enshrine things like DVT and liquid staking and it will generate the highest share of rewards and it will as usually it will provide the economic security and will be slash so that's similar to the things uh validator doing today but on the other hand you now also have the light services and they could provide different things from the heavy services so things that are not that Hardware that have not have not that um High Hardware requirements for example um censorship resistance so you could relay transactions uh or inclusion list for better me protection um and the whole thing will not be slash or will only have very light penalties so you do not need to stay online all the time and potentially in the future it could run run on a mobile phone or something like that so that's a pretty interesting concept and on top of this it also brings the concept of delegators to the ethereum blockchain so um that you can that you actively have a participant group that can delegate Capital uh to operators that then will do the task so that's also a novel concept for the ethereum ecosystem so that's something other chains had already like Cosmos for example you can do delegations there to validation notes um as of today but I think it's pretty cool because now every participants and every group in this ecosystem could find a place in the ethereum consensus so it I think it's a pretty neat idea and we will see how this turns out so that's definitely something for the long term and not completely fixed yet but yeah a pretty interesting concept and if you want to read more about it I can highly recommend this QR code all right so the conclusion uh key points um there are exciting stuff happening especially in the staking ecosystem uh DVT is in my opinion a new driving force and it will Ex will be very interesting if the max EB the maximum Effectiveness balance will go up and there are there might be validators with up to 2,480 e stake so I think it might be a good idea to have something like DVT in place when you stake like the 64 times the amount you can stake today on a validator and I think it's also a pretty neat idea to have that staking uh with uh with your friends with DVT so to to just sign up and use your friends and uh run a valid dator together restating and irland especially is pretty interesting as a completely new concept that adds adds outside things to ethereum and what's also pretty interesting is the compet ition because currently there's eying layer and there are no other big players currently as far as I know and it will be interesting if the um yeah if their competion will spin up in the future and rainbow staking in the long term might be a pretty cool um idea or pretty cool proposal to actually widen the um efolder participation in staking and all the key points in my opinion could lead to a more decentralized and neutral ethereum and that's amazing and that was my talk and if you have questions feel free to ask them or you can also join our Discord and write me a message there that's [Applause] it all right any questions the you you go back um if you're operator um is there is there a reasonable way you think you can diversify the risks of your abs portfolio in the same way that um you know the LDS are diversifying this with multiple yeah so i' I there is was a lot of discussion ongoing about this so one thing could be for example that you spin up one operator for a specific AVS and then you can let the user actually choose which one he or she wants to delegate to but one issue there is that as of today there are qus you need to fulfill to be an active operator in a set of an AVS so you need like for example you need like 2,000 or 200 um e restate on your operator to be active in the service and that might be not possible then so as of now it would most likely be a discussion point of the operator so for us if we trust the service if we like it if we want to support it and then we will go with it do you think the number of services will just increase also to a point where um there so much St in the the icon part that you're so Diversified acrosss one sort of probably yeah that that yeah that could be a possible way yeah I mean if you have like 2,000 abss connected then you have a very Diversified portfolio and that could be also a way yeah I'm just curious a bit about business you you mentioned earlier like the me part that's also part of the the whole business and Ma he mentioned that there's like streak per of Revenue basically on the on the invested State and you said like yeah 10,000 validators right and 32 E Yeah by by 10,000 so is that do you have a picture like is is that where's that basic business model is it is it just the staking is it the m is there a setup of that get a picture or is more more because to add even more on top Yeah so basically our our Revenue model is pretty simple uh you stake with us and we take a commission of your staking reward was that's it basically and everything on top like the IG layer stuff that would be in the future something interesting how we can get a share of revenue from that as well but yeah it's just the commission we take from you and that's it pretty simple and if you buy light St you're actually pay pay a little bit light yeah yeah it's like 5% for Lio and five for us I I think yeah is there anyone against the maximum effective balance or is this question quick that's an interesting point I think there are still open some open discussion points on how the slashing will happen to such a validator uh vitalic just recently um had dropped a new post about how you could panelize um the current set of 32 e validator that are running on one machine because if you see on a network interface or on a network view that a bench of Val datas goes down at the same time they are probably running on the same machine and you could penalize them more than just one single solo Staker uh who has running like one on his machine have a short downtime and I think with this missing piece you could actually have an yeah discussion starting point how you would like to to P penalize or slash the big um valid datas then okay cool thank you