# Validating Made Light and Simple

- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2023-10-09
- Duration: 28:13
- Topics: blockchain, smart contracts, distributed ledger, ethereum, Science & Technology
- Watch: https://streameth.org/watch/yt-kX-Q2fiLecY
- YouTube: https://www.youtube.com/watch?v=kX-Q2fiLecY

## Description

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on IPFS and more.
https://archive.devcon.org/archive/watch/6/validating-made-light-and-simple/index

The network has seen an increase in staked Eth led by staking pools, liquid staking protocols, CEX and solo stakers. The complexity of running multiple clients (Execution, Consensus, Beacon and possibly MEV) means that validating is increasingly considered an organised activity for teams of specialists. 
Nimbus is pushing for simplification to ensure it remains accessible to individuals and institutions alike. In this talk, Nimbus presents its vision for the future of Ethereum client experience

Speaker(s): Kaushal
Skill level: Beginner
Track: Staking & Validator Experience
Keywords: Simplified,client,experience

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon 6 was held in Bogotá, Colombia on Oct 11 - 14, 2022.
Devcon is organized and presented by the Ethereum Foundation, with the support of our sponsors. To find out more, please visit https://ethereum.foundation/

## Transcript

foreign [Music] good afternoon everyone my name is koshal I am from the Nimbus client team hopefully you'll know what members client team is but if not I'll talk to you about that I support Nimbus client team with their ecosystem development efforts which means I get to go out and talk to end users who use the Nimbus clients or other clients for that matter and understand what their experience is today and what their expectations are of the future so in this talk today I should start to use the clicker as well actually foreign there we go yeah so I just wanted to share some of my thoughts and experience talking to a lot of the validators right whether they are centralized exchanges whether they are the professional node operators or individual users I just want to share the feedback that you get when you talk to these validators about what's the future of the client experience like what does the client experience need to look like in the future as you can see I named I titled my talk validating Made Simple light and simple it may be light but it's definitely not simple right now so that's something that we definitely want to fix in the future let's talk about Nimbus a little bit so Nimbus is a ethereum client it's a consensus client that you can use right now it's an easy efficient uh low resource intensive or less very efficient essentially um stable consensus client so you can use the Nimbus client with any other execution client if you're running an ethereum node um the story behind Embassy is obviously around lowering the barriers to enable participation in the ethereum network and we are very specific about the words it's it's the ethereum network not ethereum in general you want more participation in the ethereum network truly see it decentralized further the reason behind that is also the next slide which is it's part of the status Network portfolio of projects you will many of you will recognize the status Network project or the status app for example many of you will have that on your app on your phones as well and hopefully you're using it but the story is status started back in 2017 with a decent with the mission to create a decentralized peer-to-peer censorship resistant chat application but as we started building it we realized that there is a lot of work on the infrastructure side that needs to be done and that includes building the infrastructure for enabling a completely decentralized peer-to-peer messaging storage and compute and that is where some of these projects actually fit in if you want to know more about some of these projects please pop by the Nimbus booth on the on Level 3 and we'll be happy to talk through the details of each of these projects but there is a Common Thread that runs across all these projects and that is to enable true true decentralization right and the focus on privacy and censorship resistance and that sort of theme flows through to Nimbus as well so when we talk about the future client experience right it needs to be in the context of who the users are and both the current users as well as what we see the composition of the you know valid data set to to look like maybe a year or two down the line or future in future in general right and who are our current and future users so our current and future users can be broadly classified into categories it's the individuals or you can call them the solo stakers right and the professional node operators now there are some other you know you might call them customers or segments that are in between but broadly speaking you can put them in these two one of these two categories and we can work through the details as we walk through the presentation so let's talk about each one of them we will talk about the professional node operators first and then we will talk about the individual of individual users and we'll see what the similarities and differences are in terms of their expectations of the of using a client to run a valid data so professional node operators like who are these let's do a quick show of hands for people who have staked their eats with let's say a centralized exchange or Lido or rocket not rocket pool but any of the Lido or centralized exchanges for example all right a few few hands here that's good that's good uh we'd hopefully see more hands when we come to the individual sex segment but professional node operators are an important part of the part of the network right now like it's not a segment we can ignore right whether we like it or not they are here they do offer an important service right now and they are a large part of the network that is that needs to be taken seriously in terms of their expectations as well so in my role I have had the opportunity to speak with about 12 to 15 node operators in the last six to eight months and it's really interesting when we talk to them about their setup the history as to why and how they have set up the node operations that we understand you know why they've chosen certain clients and why they've chosen to work with certain clients and in that process we learned a few things which is which is really interesting some of these node operators in fact most of these node operators are quite large right by ordinary means I mean in terms of comparison they run anything between two and a half two seven eight thousand validators that is a lot of validators right and these These are individual node operators maybe 20 30 of them I don't have the exact count but I suppose that's somewhere close to 30 35 node operators total across some of these liquids taking protocols in general um they tend to operate from specific regions they generally are a registered entity in these regions which means they are governed by the law of those jurisdictions and they're likely that they're trying to obviously scale up through the liquid striking protocols but also directly trying to find customers who would like to stick with them right and the other aspect that is really interesting is that they're trying to also serve the needs of Institutions operating from the same jurisdictions why because they understand the jurisdiction they understand the regulatory aspects and the and the nature of business in that in that jurisdiction and the institutions in those jurisdictions seem to have a a inclination to use and you know work with node operators from that region right so it kind of makes sense that they are sort of becoming a channel for the institutions to come on board so that's another Point another reason why node operations and node operators in general as a segment are an important segment of the network and they need to be looked at in in that regard um let's look at what the node operators actually look at in terms of performance or what they prioritize as the characteristics that are important the first is the uptime right now when you're running 7000 or 8000 validators uptime is really important you can't be down for hours it needs to be up and running there needs to be fair degree of failover redundancy in the system it's like a it's a it's a significantly professional or sophisticated operation that the node operators actually run the second being reward performance which is extremely important so the choice of client is also based on you know is the reward performance good enough um and finally operating efficiency like you don't want to hire too many people to run the operations as well because that impacts the profitability of their of their operation in a way so these are some other the fact the fact node operators think about when they think about their business in a way and this is important because this feeds down to what they look as in the client when they select a client right so things like stability of the client is extremely important things like reward performance how does the reward performance of a client a compare to client B these are factors that are extremely important uh to node operators factors like fit with the current setup this is something that is extremely important why because each client is slightly different and this goes back to Danny Ryan's presentation on day one where he talked about diversity of perspectives and each client team takes a slightly different perspective on how they would like to architect the client now when it comes to node operations node operations like similarity homogeneity in the client architecture so that it fits with the way they have started to operate right so it's really interesting that the uh there's a huge push to try and figure out how to make the clients actually fit and work with the with the current architecture of the node operator and the future and in the direction that the direction in which this architecture is actually evolving other factors are are really important such as release communication and the Rhythm so how good can the quality of release communication the rhythm of the release communication and the ability to contact the team the ability for the team to turn around when there are issues and challenges and of course the testing rigor and the variety of testing that the team actually conducts before the release scores are so these are all very important factors from a node operator point of view right and as a client team we have to take these into account as expectations of a potential large set of customers right who use these clients essentially now what does desired client Evolution look like for from a node operator point of view it's basically much more flexibility in how they would want to use the client to make it work with their current setup and much more interoperability interoperability because it's not just the client software that is used there are other other pieces of components of software that are actually used in running a sophisticated node operation right so interoperability with all of these components becomes really important when it comes to selecting a client and working with a client all of these factors on all these discussions are really important because if we want client diversity in the network then we have to somehow make sure that we meet these requirements we meet these challenges so that we can see greater diversity in this segment of of validators which is primarily the node operators which we are talking about here so a little bit about what Nimbus is trying to do to address these issues right so nimba started when it went into production in 2021 it's it was constructed as a very simple easy client to use it had a single process a beacon node setup it had validated duties that are performed by the beacon mode itself which means you don't need to go and set up a separate validator client so this is really easy for anybody sticking from home however as we've talked about it previously a lot of node operators actually started using some of the other clients which are architected slightly differently which means a separate Beacon node separate validated process and the ability to mix and match between these components so it is an expectation that Nimbus also offers something similar if if Nimbus is to be used um equally well in this segment of validators in this segment of customers right so there are a few things that Nimbus has done over the last I'd say six to eight months things like supporting the Nimbus Beacon node with support for Welch and the lighthouse client which we have now in place and the Nimbus web3 sign or remote interface remote standard interface as well which is in place so these are things that are really important for node operators not as much for somebody who runs let's say a single validator from the home but these are expectations from from a node operator perspective the one big requirement that was missing the missing piece I suppose was the Nimbus validator client or the ability to split the client so having the ability to use a separate Beacon node and the validators separately and use them into you know use them with other mix mixing and matching it with other client components essentially Nimbus did not have that so far but we have it now and it's available in beta it's going out of beta into production into and ready for mainnet very soon so these are all the all the components that have been built over the last few months last few last six to eight months to enable nimbers to be better suited for this large segment which is the node operator segment um that is not it though there are some Advanced features which are which Nimbus is working on some of the validator client features that are really interesting for node operators and I'm talking about these because these are factors that are important to node operator segment which need to be taken into account when we think about the future client experience as it evolves because the client is the one client that is used by both individual users and the node operators at a professional level so the advanced validator features are things like they are inspired by the secret shared secret validator research and IT addresses risks like Rogue employees malicious actors or infrastructure failures even in the current setup where you have validators running through web remote signers infrastructure failures can still bring down the operation in a way so that is one area of or risk potential risk that the the node operators are still exposed to there's always the risk of Rogue employees because if an employee has access to all the key keys they can actually you know hold the company to Ransom right and you don't want that when you're running 7000 validators so these are some of the factors that are really important for for to address from a risk management point of view there are some features that Nimbus is working on in terms of what the advanced validator feature set would look like that would address some of these some of these issues and part of this is about splitting the sysadmin teams into multiple teams that have only access to part of the keys like no but no one person has access to all of the the entire key set I suppose and distributing that the portions of those keys across several different infrastructure providers potentially the challenge also here is that this is extremely complex I mean node operations itself is complex and then on top of that we are we are overlaying organizational changes that you need to bring about so this creates an additional layer of complexity but it's a trade-off right it's a trade-off around do we want to actively manage that risk versus sort of live with this risk as a node operator and in this regard the there are a few possible solutions that Nimbus is exploring including kind of a new remote key store where you know each item is sent to multiple web3 signer instances instead of just one so there are some of these things that Nimbus is exploring which will add much more capability to the validator client which is what the node operators have been looking for for a long time all you can see where the expectations of node operators have the expectations of node operators are changing with regards to the use of the client right and how they see the evolution of the client experience now this is definitely not what individual customers want right like as an individual if you want to run the node you're looking for Simplicity and so it's almost like pulling the um pulling the direction in which the client should evolve in a different direction totally right compared to where the node operators are seeking for the client to evolve question before I I can't go back unfortunately but I I just quick show of hands for those who are running a note from their own homes wow that's that's definitely more than people who are staking through liquids taking protocols or or exchanges so that's that's really good to see um Hester and Kiel from the team ran some research very recently and the feedback is quite interesting from individuals there's a lot of potential there's a lot of interest but they're all held back by a number of challenges particularly around lack of easy access to documentation lack of you know uniform documentation multiple clients have multiple guides you need to shuffle across these things it is not easy multiple Discord channels for example so there are a few requests that came out when when there are in this this research right one of the first is is there a no command line option please setting up a valid data right believe it or not they all the validators you set up right now no matter which client it is it's all still command line right so a lot of users end users who are interested in running validators were not very tech savvy are still looking for a non-command line interface to set up the validators a fair ask right um the Second Challenge obviously is don't have the 32 or 16 age so how do I participate right so this is huge idea in the in the in the in this segment of individual users which is around that you need 16 to 32 weeks to participate in the network but there could be other ways to participate in the network when you're not validating and we'd like to explore a few of those in this talk as well and finally as I said earlier the information is scattered across forums across across different guides across guides that are created by champions in the community and then Discord channels so it becomes difficult if not you know impossible to coordinate and to be across all of this especially when there are multiple releases going out for the client um professional node operators have the bandwidth the capacity to be able to deal with this individuals don't right so there's an absolute need and I know the community has been asking for individuals to step up and take on uh the responsibility to run more nodes run it run them at home but there are real challenges and we from a client team perspective we definitely acknowledge that these are challenges that need to be addressed like let's give credit where it's due right so credit to Rocket pool to uh to uh stereium to DAP node who have all stepped up in a way to offer some solutions while there is a gap tap node Stadium offer ability for you to spin up your own nodes right on your own Hardware with a UI so you know you don't don't need to necessarily use a command line interface so these are these are absolutely positives and there's an option to use Nimbus across all of these uh all of these operators rocket pool of course very popular amongst people who want to run a node from their own homes whether that's a full node whether that's a mini pool and again Nimbus is extremely popular as a client of choice in amongst the rocket pool users uh limbus is doing its own bit to contribute to this effort right so we completely acknowledge that our challenges here that need to be addressed and we have some views on how this can be addressed the first thing is abstracting the complexity away from the end user Right End users like to be taken on a journey if you can give them a wizard and follow the journey follow the wizard to get to set up a validator successfully easily and be able to monitor that in a simplistic fashion then that is what the end user would like as an individual so one of the key tenets here from a client Evolution point of view for individuals is abstracting complexity away from the individual and this takes the form as far as status is concerned we've talked about let's say theorem and death node but status is heading in the similar Direction the status Network team is actually planning to integrate the Nimbus client inside the status desktop which will give users the ability to spin up a node manage it through the status desktop application so this is one of the things that Nimbus and the status team are working on to address second one let's talk about the elephant in the room who's recently set up a note quick show of hand all right um You had to set up one execution client consensus client or probably even Beacon node and a validator client let's talk about the two components individually and then possibly an Mev setup right that is like four steps it's too complex right for individuals when we're talking about no command line options and presenting a simpler UI walking through setting up four different clients with four different guides but potentially all components at least is complex and this could get even more complex if it's not checked right now so we need to do something about this and from a nimbus point of view one of the things we would like to do and we are working on is a unified execution plus consensus client or actually an execution client that works really well with the consensus client from Nimbus so a nimbus execution client and a numbers consensus client together um there will be ability to use these clients independently and interchangeably with other clients as well but I think there is real benefits if we can simplify the setup of our of a validator and take the user on a journey where you set up your execution client and consensus client simply in the same process rather having to switch between things there are other ways to participate in the network so the light proxy this is something I want to spend a bit of time on and I think I'm running out of time but I have enough time to cover this as well so the light web proxy is something that Nimbus has recently released and this is another way to participate in the network um participating in the network doesn't only mean a testing and validating and creating blocks right participating in the network also means taking responsibility and checking verifying you know where you're getting your data from can you trust that data should you trust that data if if you should if you should not trust the data you should at least verify it right and so lightweb proxy gives you the ability to do that it's a it's a lot it's a you can basically set up the lightweight proxy on your phone right and instead of trusting the endpoint that let's say the wallet trusts the endpoint that's set up for the wallet instead of trusting the data that's provided by the endpoint the light web proxy can actually intercept that call to give you an example if you make a call you want to check your balance on the on the wallet the wallet makes a call you know get balanced get it balanced the lightweight proxy can actually intercept that and the light web proxy connects to the light client Network to keep track of the head of the chained right now or the state right and then it and then then it requests a get proof separately from any of your not providers API providers to check the state route against the balance that is supplied so now if let's say infuria you use metamask within fura and if inferior comes back with a balance your light web proxy can actually verify whether it is true or not whether whether that balance is correct or not so there are all these things that that can be done to improve the security and health of the network that don't only mean validating and staking there are things that individuals can do on their own to improve the the secure their own security the security and and Trust in the in the balance that they see in the metamask wallet and these are things that are possible now right now so on that note I would also like to say my colleague Eaton is going to present about the like Clans tomorrow so if you are interested in in the like client and the like client proxy please do make sure you you attend his talk it's quite fascinating to see the actual demo as well if you can show that and finally in conclusion I'd say the client experience or evolution of the client experience is being pulled in different directions right more modularity when it comes to node operators and simplicity when it comes to individuals so it's the task for client teams in general isn't easy to balance these priorities but I think I can speak on behalf of nimbus that I think for from a nimbus point of view there are certain things that we have planned which should hopefully address a number of these things for both these major segments of validators because both of them are equally important at this stage on that note I'll conclude I'm happy to stay in touch my Twitter is lift lines and you can also reach out to me via Telegram and other other channels using the same handle thank you [Applause] any questions second row oh today you know you're close enough you could probably yell if you want hello hello thanks for the sharing um I'm just curious um for the better user experience were you like thinking about provide infrastructure skull to set up the the infra like telephone or Hampshire this kind of thing so if I understand correctly I think are you asking whether or not um numbers would Supply infrastructure is that is that what you're asking sorry infrastructure is code right I think the primary goal for us at the moment is to make sure that we can take the boxes that we have planned on the roadmap right so these are these are things that we must address over the next six to eight months um um it could be one of the things we could consider absolutely but uh yeah I wouldn't say more about that than that at the moment I think for us it's important that um the users who end up using Nimbus whether that's directly using Nimbus as a validator that they set up on their own home environment whatever that looks like or use rocket pool Stadium dap node or any other Hardware providers to set up the client or have you know all used as part of professional node operations at scale that they have the flexibility to run it the way they like it right so I don't think we'd like to mandate anything in particular but the guide does cover different scenarios um depending on who who wants to set up in different ways great thank you let's uh [Applause]