BlockTalks with Darius Dovydavičius

[Music] In the 225th block talks, we speak with Derivost David Aviču's Web 3 Partnerships Manager at Sherry Servers, who talks about the harder layer of blockchains. I'm your host, Mauriz Uma Gauji, and this is Block Drops, that weekly digest on blockchains for business. These news are not a form of endorsement, sponsorship, or encouragement for consumption, and are meant for educational purposes only. [Music] All right, everybody. I'm here with Derivost David Aviču's, who's going to tell us who he is, and what he does with Blockchains. Thank you, Maurizio. It's lovely to be here. So, yeah, I'm Derivost. I work as Web 3 Partnerships Lead at Sherry Servers, and Sherry Servers is a bare metal provider. So, I can set a bit of a stage for anyone listening who is not that familiar with hardware. So, bare metal is just a fancy name for a dedicated server, and a dedicated server is a computer that you don't share with anybody else. It usually sits in a data center, and a data center is basically a modern warehouse that has redundant power supplies, multiple internet providers, and is physically secured. So, it kind of is built to keep your servers up 24/7 no matter what. Okay, so getting back to Sherry Servers. So, as I said, we provide bare metal infrastructure, and this means that we focus on keeping Web 3 workloads, running on secure, fast, and robust infrastructure. And me, personally, so my job is pretty simple. I just talk to people. So, I listen to people's needs. Like, I talk to Web 3 business owners, what they need, what kind of hardware they need, and try to match that with the stock that we have, basically. And another part of my job is to gather feedback from our customers, like, what they need, what setups they might need in the future, what features are not there with the Sherry servers, and transfer all that information to our DevOps team. So, they would have something to work on and to build. So, yeah, I'm basically just a translator. I sit in between customers and and and and. And that's pretty cool. Now, I do have to ask, and I know people from the audience will have the same question, which is, you kind of described the cloud, which is someone else's computer. What are the differences you came point, because we know for a fact that most of Web 3 and Blockchain run on cloud providers, be it Google, who's just recently announced their own blockchain, AWS, and even IBM now has a partnership with with the DFS. So, these are all big providers, and they all operate data centers, and they offer cloud infrastructure. So, for the non-initiated, what would be the core differences between what you do with Sherry servers in terms of the bare metal offering versus a cloud offering, or one of those big tech giants? This is a very legit question, and it's not Web 3 specific. So, the main difference between the metal providers and the hyperscalers, like you mentioned those, is there is no virtualization layer on your hardware. So, this is the main difference. I mean, when you go with the metal providers, you get full control over your infrastructure. Nobody else is using that specific server. So, you have full CPU, full RAM, and all the NVMe drives you have in your computer. Nobody else is building on the very same machine, and you are sure what you are paying for. So, that's the main differentiator, and getting back to Web 3 and Blockchain, it really depends on your use case. There are chains that are really fine running on clouds, and there are some chains that really need this specific powerful machine that's only where only your own workload is running. I can appreciate that. Being a builder myself, there are some trade-offs and some caveats you have to take into account when you're choosing your hardware infrastructure, and that is an interesting thing. But let's go back to your origin story. Did you start in the Blockchain industry? Do you come from a previous industry? How did Blockchain or crypto or Web 3 enter your life? OK, so I think that would take me to a bit of a story behind cherry servers. We have been around, we like cherry servers have been around for 25 years now, so we are, like usually Web 3 people are really surprised when they hear this. It's not that we are the recent Web 3 only company that just popped up during the last bull run, so we've been around for quite a while, and I've been with the company for the last 10 years, I would say. So during all that time, we've had all kinds of different customers on our servers, gaming, streaming, rendering, business systems, databases, just hosting websites, and so on and so on and so on. But like three years ago, yeah, I would say around three years ago, we started noticing kind of a different customer profile, they needed beefy machines with high CPU cores, fast NVMe drives, and lots of bandwidth. So this wasn't something that we couldn't provide, but we just were like, what's going on? Like, why do you guys need all these machines? Like, what are you building? Yeah, so we started talking to these customers, and this is how we learned about Web 3 space. It appears that all these people who needed all these specific machines were running validator nodes or RPC nodes, and most of them were from Solana space. So this is how we were introduced into that. And yeah, so we as a company gradually shifted, like moved to faster CPUs, more fine tuned RAM options. We bought a bunch of new faster and larger NVMe drives, and all this change happened not overnight. It was lots of small adjustments on our side, and we did this specifically with the help of our customers. We spent hours of listening and trying to understand what are the actual pain points on their side, like what they would ideally like to get from their server provider. And those customers were really generous with their time and their knowledge, and they shared all these things with us, and yeah, we are where we are now just because of the guys who actually kind of took a bet on us when we were still trying to figure out things in this Web 3 space. And yeah, they helped us a lot, and we were really grateful for that. I wanted to book click on that because I'm also a product person, so it echoes to what I like to do and how I work. But I would I'm obsessed with use cases when I when I talk to people about blockchain on the show, I tend to gravitate more to use cases because it helps me convey the power of the technology. This is what blockchain solves uniquely and this blockchain is different than that because they solve different problems in different ways, and it makes it more relatable to people. We're going quite a notch deeper into the hardware layer of blockchains, right, with with the conversation with you. What are use cases in your language? And what were the use cases that most emblematically helped shape cherry services, servers offerings as you just laid out? Can you pinpoint like two or three use cases that really made a point of, okay, we need to go into that direction because these are the use cases that make sense for us to support. I think it was Salana people that made us make the most changes on our side, like Salana being one of the fastest chains also is one of the most demanding in terms of hardware and networking. So we kind of figured out that if we solve Salana use case, we should be fine with all the other stuff. So yeah, it was a bumpy road on our side, like Salana needs lots of bandwidth with occasional spikes, and it was something that we really needed to sort out on our side. Like we had enough bandwidth, but we didn't know how to handle the spikes. So we needed more and broader internet lines, better connectivity, more internet providers, and so yeah, so that was the networking part, and then the hardware part. We really needed to upgrade our stock to buy more faster CPUs, and yeah, I already told that, so I think that we kind of solve this, but you know, it's a moving target. You can never be sure that you did it. You can never say that, hey, we're done. Now we can just get some rest. The requirements change every week, like new updates and new upgrades come in, and you need to adjust yourself, and you need to constantly monitor what's going on. So it's another ending game of improving one of the things that I covered a lot on the show is the interface or the intersection between Web 3 and crypto and AI, and we know how AI has been in the headlines, especially when it comes to the hardware demand and energy and, you know, water for cooling down and all the other stuff in this new kind of generation of GPUs and all these new racks coming out and everything else. Has, have you seen the uptake on that on the business that you're running, and have you seen that also in this same intersection, is this something that say because you guys are so focused on Web 3 and all the industries, you might be better position to potentially interface AI and blockchain in a single kind of stack. What's the take there in what have you seen in this emerging interface? We hear a lot about AI. It's kind of a buzz word right now, and it's been like for a year or two right now. But we deliberately made choice not to go that path. It's a completely different business model. We decided that we are a high density CPU provider, and we chose to work with workloads that are CPU specific, so we just don't want to go the GPU path. One of the reasons is it's much more capital intensive. Another reason is that we kind of tested the waters back in the day and tried to try to understand if this might be a viable business model for us. But it was really hard to get our hands on GPUs at that point, because it was a huge demand, and everybody needed GPUs, and we were not the first in line with the providers and with the producer. We just see how decisions can be both strategic and opportunistic, and some people have the clarity to know when to say no, really. It's one of those things that are kind of make or break a business, especially when these kind of inflection points when there's like a herd mentality moving towards a direction, and you're like, no, I'm going to stick to my strengths, and I know that I can do well with what I know. I know what the domain expertise that I have, and I feel that, you know, that's there's a very interesting people would say that, but if you have established business and you have the data and you have the customers in your partner with the customers to consistently meet their expectations, I think there's a lot to be said of know yourself and then make these plays in ways that can just sustain your business in the longer term. I really admire the posture and the attitude there and interesting take. Now, let's go back to blockchains and web 3 in general. Can you specifically on on those cases, you mentioned Solana as one large player in your ecosystem, what are the other ecosystem projects that you've supported over the years, and are there any specific use cases that were other than the capacity and performance required by Solana, are there any others that are curious or peculiar that have come across over the years? We do not run any nodes ourselves as an entity. We just provide hardware, so this is a very important point. We have customers that run their nodes with cherry servers and we don't want to compete with them running nodes ourselves, so that's one thing. Let me know for sure that our customers run lots of chains on our servers. So it's Solana, it's Sui, Aptos, Polygon, Monad, and there are some names that I can't disclose due to NDAs, but one of the recent interesting moments was the Midnight Network scavenger Mint. I'm sure you know that. Like a couple of weeks ago, there was scavenger Mine event on Midnight Network where basically anybody could just go ahead and mine Midnight Network tokens on their computers or on their laptop, so I'm not sure, maybe even on the phones. And this is an interesting point because there were some professional miners who spot the opportunity to do it like in a larger scale. And they used our servers to run a bunch of minor machines and at some point we were completely sold out. Like every single server on cherry servers has been out of stock exactly for that. So this was a nice stress test for us to see how everything works and do we have enough capacity to actually power up every single server on our racks. We did well and now the scavenger Mine period has ended, so we are back to normal. But yeah, that's just one example of and it shows like that cherry servers is the well known name in the Web 3 industry and people know that we provide good reliable services and yeah, they can use us for whatever workloads they need. Do you as a professional have you onboarded clients or friends into Web 3 and how how do those conversations tend to happen? Are you someone who's evangelizing on Web 3 or bringing people to crypto? So how does these conversations happen if they if they do mostly know I've had these kind of conversations before, but they usually didn't end very well and so now when I talk about crypto, I always say that, hey, this is not a financial advice. And if you're going to invest into any cryptocurrency, invest the money that you are not afraid to lose. So that's my only take, but I usually don't talk about crypto with my friends and family and I don't evangelize. But having said that, but having, yeah, having said that, I really think that Web 3 as an industry and as a field really has the future. It's just so volatile that I wouldn't dare recommending or anybody to put any money into that. So let's talk about that a little bit because it's an industry information right the regulatory frameworks are coming out just now. We're learning what good looks like. Obviously, there's a lot of back history and financial services that we can lean on, but clarity is coming at the moment. As an industry, what do you think is the one thing we need to fix to avoid working another 20 years here and then realizing we didn't fix that one thing and now we're back to square one. What's the one thing and why I think we need to define rules like clear regulations like if we have rules and everybody in the space know how to play. And this is the, this is a single point where we really need improvement chair servers is based in in Europe. We are in a very beautiful small country of Lithuania and yeah, so we have European law here and the micro regulations kind of does that, but it's still very vague and very. At this stage, so I think the faster we get things done with the regulations, the faster the industry can move forward. Let me double click on the on the Mika commentary. You said it's opaque. What would make it better in your opinion, given that you're in Europe, you're regulated by European laws and Mika is kind of brand new. So what would make it better? I'm quite sure people who actually wrote all this regulation and all this documents have really touched crypto at some point. Maybe some of them did, I don't know, but it would be really great if they would have more knowledge about that. Sometimes it feels that it's been written by just people who are really good theoretics and and don't have anything to do with the real world situation. But that just might take and it doesn't necessarily tells what your servers as a entity has to say. So let's separate those things here. We can kind of summarize this as an invitation for regulators and legislators to actually come and use the infrastructure they're trying to regulate. I've been saying this for quite a few years, we welcome regulators. Let's help educate you so you can write not more regulation or less of regulation, just better regulation for this new paradigm and allow a market to form around the technology, which is all we've been wanting for, I don't know, a decade and a half now. So quite a few years for them to pick up the baton. In terms of an industry that doesn't sleep, right, you're on a hardware industry, it's even worse because they just machines. How do you keep yourself sane and healthy in such an industry? I heard on one of your talks earlier that you said that you don't understand people that do running and you think that they are crazy. What I said is I hate running. So I do run, I run. So there are two states like in my life, either I'm sitting in front of my computer and work or I run. So in the days where I don't run, I usually make like 2000 steps and it's just getting around around my home and when I do run, then yeah, it's several thousand meters. But yeah, it helps a lot. Let's cycle path. Okay, I get you, I get you. No, I'm joking. I have short legs and I'm heavy. So that's why I hate running, but I appreciate people who do run. And I know how, how great it can be overall for, you know, bone density, muscle, cardio and all that stuff. So it's more of a, it's more of a stab than anything else, but yeah, I wish I could run better. That's the joke. So before we finish, as we start the year, what are the three things you're most looking forward in 2026? I would say it's just one thing. I'm looking forward to continually improving our service. Really, like we have so much stuff to improve, like add more locations, add more CPU models, add more networking providers. So it's, it's another ending story. And I feel that lots of people in our company love doing that. So yeah, I just want to, to be there when, when the year ends and say, Hey, we did a good job. Let's move on. Here's to moving on. And how can people learn more about you and the work you do at Cherry? I think Twitter account and LinkedIn and yeah, and for Cherry servers, just go to Cherry servers.com and you will find everything there. Awesome. There is. This was an interesting conversation. I think it's the first hardware person that we get on the show. So I learned a lot from you. Thank you so much. And maybe I'll bump into you in a crypto conference somewhere. I'm sure we'll meet at some point. Thank you. Yeah, thank you. Catch us online. We're on Instagram at block drops podcast on Twitter, at block drops pod or zero ex Mauricio. We're on lens at block drops dot lens. We have a newsletter on LinkedIn. Right to us at block drops podcast at gmail.com. And you can listen to the block drops podcast at Spotify, IcoLab, Faberbon Tech, and all of the other major streaming platforms. Yay! Shout out to Darius for taking us into the depths of the hardware for blockchains. And shout out to Serene Melican for making this conversation possible. Don't forget to leave the ratings in a favorite player and check out the merch store at block drops podcast dot xyz. This is all for today. Stay rare. Stay weird. Have a trip.