Parallel Processing and Parallel Execution for EVM chains

Transcript

All right. Hello everybody. How you all doing? Got it. Yeah. So, uh, it's great to be here. Um, my name is Stuart. I'm the CEO and co-founder of Cadenna. And today we're talking about parallel processing and execution on the EVM. So, so I'm from Cadenna. Cadena is a uh proofof work layer 1 blockchain that offers a unique architecture uh with parallel chains for scalability. And I'll be telling you all about that uh in our talk today. But uh historically we

never had EVM chains. So what we're talking about now is a is a new way to scale applications for EVM technology. Um, a little bit about me. I've been in technology uh since the 90s. started at Apple, uh came to New York, got into fintech, uh trading tech and uh specifically ended up at JP Morgan, uh was one of the founding directors of JP Morgan's blockchain group in the early 2010s before, uh where I met my co-founder, Will Martino, and we uh founded Kadana coming out of JP Morgan in 2016.

So today uh we're going to talk about a topic that everyone talks about a lot which is you know how to scale EVM technologies. This is kind of the main problem people are dealing with in most respects. So um so the main approach for scaling in EVM as everyone knows is layer 2 uh or layer three. And uh so before we uh jump into it, one thing I I want to point out is that in software engineering scaling means you have resources that you can throw at a problem when load gets too high. Ideally it's a uniform resource. So for instance on AWS that might be more RAM, more GPUs, more uh more compute and no matter

how much load you get, there's always more you can grab. Um so whereas with L2 uh that you know to there's a lot to like there because there's certainly an almost I mean there's there's a limit on how much can be uh cleared on Ethereum but that number is going up over time. So there's a lot to like however there we are losing something with layer 2 and so let's talk about that. Um so most layer 2s have you know basically this uh structure radically simplified of course which is you have the layer 2 you have the something called a sequencer which is the kind of critical bit that makes it possible for transactions to be uh you know to

execute and then record on the layer one which is usually ethereum. Um so let's let's see how that's going. So one, so while layer 2s give you more throughput, uh, finality has quite a bit of latency. If you want these to show up back on the main chain, it can take as much as a week. Um, settling back to the chain can involve significant gas price. Um, sometimes just depending on how many uh, rollups are settling at the time you settle back. And actually many L2s are also uh hitting their limits as well uh because

most of them are very similar architecturally to Ethereum which means they're they can only do as many transactions as as they have a hard limit and once you hit that gas prices go up on the layer 2 as well. Um but perhaps the most concerning thing is the sequencer technology itself which is inherently centralized. Uh it's you know blockchain it taking a step back blockchain was built on proven technologies. Uh the Bitcoin white paper introduced no new technologies. It just put things together in a novel way. Uh same with the V1 of Ethereum. So when people talk about decentralized sequencers this is technology that does

not exist yet. So and it's possibly not even solvable. Um so another issue is the use of optimistic protocols uh in layer 2 um which basically assume that when things get back to the base chain there won't be conflicts uh for a given account. um if there is a conflict uh this can lead to delays because you have to retry and wait and so there's there's a fundamental kind of state management problem with layer 2 that uh you know there's a happy path and lots of you know things like NFTTS and things like that can go down the happy path but there is this way that it can slow also get slowed up um and under load this

only gets worse um these conflicts become more frequent and and cause uh repeated roll backs. Um and just a quick sidebar is that some layer 2s are now offering parallel execution. So that's interesting in the sense that they uh overcome the problems we were discussing earlier where layer 2s uh can have gas spikes of their own. So there's faster processing uh more capacity. However, uh the state contention problem doesn't go away and arguably it gets worse because you have the potential for many more uh conflicts to occur once you get back to the base layer 1.

So what we're introducing with EVM technologies based on our architecture that at Cadenna we've been running since 2020. So I want to introduce you to chain web and uh so you can understand how our proofof work multi-chain technology works. Uh go back to the graphic. Oh no graphic. Okay. Uh what's next? Yeah. Okay. So this gives you an idea of uh the architecture of chain web and you know so what we have here is our native uh application layer which is called packed um uh we launched with 10 of these a year later we scaled to 20 um

and so EVM is our current new effort and we have a big announcement today so it's good you're here uh and that's going to have a similar architecture and this is all one network, one consensus that manages all these chains. Uh, next slide. Yeah. So, um, so multiple chains means that there's no single queue. So, this is great for transactions getting into the network and being processed. Um it also means that uh there's uh you know since we're talking about 20 chains intact and many EVM chains coming to our network uh this is also a solution for gas because um

you can scale across different chains and avoid congestion that way. Uh generally speaking on our network we have gas prices in the fractions of a penny. Um the other thing about our architecture that's unique is that you can keep adding chains and I'll explain a little later why that's true. But every time we add chains uh the throughput uh goes up linearly for that amount. So for instance when we went from 10 to 20 chains uh our throughput doubled.

Um so a word about our consensus. Our consensus is based on proof of work and again we'll get to some of the details about that in a second. Uh but the the notes here the kind of headline is that um the proof of work is still the most secure and the most decentralized consensus technology out there. Uh and so we use the security that you know secures Bitcoin for uh truly decentralized tamperresistant validation but with high throughput. Um it eliminates the need for sequencers because uh you just submit to the various memp pools um and you know and you enjoy the security that comes from having thousands of uh unpermissioned

miners uh mining the network. Finally uh it's a much simpler architecture. There's no middleman or anything like that. So this reduces the attack surface for uh you know for exploits and it it's a far more secure architecture. So from the uh application point of view and protocols point of view um uh chain web uses uh the chain web application layers use a native burn and mint protocol. So a given token actually exists on all chains at the same time and it means you don't have any kind of liquidity fragmentation. When you move a coin it just shows up there and it's it's like it's a single uh it operates

as a single token. Um so no need for bridges, no need for anything custodial when you're scaling things across all these chains. Um and there's no attack vector there as a result. Um and for developers, this is all native to the chain web protocol. Uh you know, so in the EVM side, there's pre-ompiles for uh system contracts for you to use to natively transfer tokens. And uh so that means any movement you do around the chain doesn't need a special audit or anything like that. It just works.

Um and this is very uh next slide. This is very uh good for DeFi and other apps to scale because that means um since uh pools pools therefore can easily scale across chains because you know tokens operate as a single token. Um and again there's no extra trust layer required. So that's uh that it it makes it possible to massively scale uh DeFi applications but in a fully decentralized and secure trustless manner.

So let's talk a little bit more about chain web before we get into our chain web EVM announcement today. Um so the idea is that chain web chain web can grow with adoption and utilization forever. Um so interestingly enough this is only possible with proof of work. Proof of work the amount of work the the hash rate the hash power that you're using can be divided pretty much infinitely um in in our architecture. Whereas if you did the same thing with proof of stake, uh proof of stake is a discrete set of validators. So if you were to split that up onto multiple chains, uh you would

lose security. Chain web has this kind of bizarre uh feature which is that as you add more chains, it actually gets more secure. And when we launched in 2020, uh we actually had gauntlet uh do a agent-based uh simulation of this and proved that the network actually gets more secure with more chains. So it's more but without needing any increase in hash rate. So the same minor mine but you can add more chains. Um and then the parallel chains act as oracles of each other's activity under the hood. And that's the reason why uh that's both the reason why it becomes more resistant to attack and why it's so easy to move

assets around the network. Um so what are the benefits here? Uh first this is very important. This is a proven technology we launched in 2020. uh again with 10 chains uh expanded to 20 and while it hasn't been a walk in the park you know it's a public blockchain out there it's a you know very lucrative target for hackers uh we're proud to say that we haven't had a single interruption in block production since 2020. So it's a pretty great track record as far as that goes and it's just to demonstrate that this scaling technology is proven and here today. Um what does it mean? means that gas stays

cheap uh because apps scale across chains and it and congestion is manageable. So the idea is that it's not like we never have gas spikes. It's just that there's something you can do. So if you deploy your app to a single chain cuz you're just getting started and then your app does really well. Sure, you can clog the chain and gas prices go up. But the good news is you can then scale to more chains and or you can go find a chain that doesn't have as much uh congestion on it. You can do load balancing. You have all the tools that uh other areas of software engineering enjoy. Um a very important point here is that you don't have to run your own chain. You can uh this is not an app

chain story. This operates as a single uh integrated layer one. So you just deploy your app and sorry oh okay um you just deploy your app and you're ready to go. So how does this impact you and this and by you I mean builders and developers and the kind of people at this great conference. Um it means um no more fighting to get into a block as we said. Uh gas stays manageable as we said no need to run your own app chain.

And an important point here is that we've been doing this since 2020. So we really understand how there's different ways you can scale and they're going to be different for every DAP. So talk to us because we're experts in how to use this technology to scale and it's going to be a little different for each application. So the big announcement uh today is that we are launching test net as of today. It is live and uh thank you It's a lot of work. The team really uh pulled one out for this. We're very excited. Um so this is the, you know, this is the big slide here, right? Um these are all our partners coming on. Um, so this is

the great time to get involved because even though we have these partners, it's still early, but you know, we have four DEX partners, five wallets, three lending protocols, two bridges on RWA market, four AI partners, and we have solutions for stable coins, GameFi, DPIN, and indexers. So, uh, test net is here. It's going to have a it's going to have a full ecosystem for you to work with. So, it's never been a better time to get involved with Cadenna. So, how can you get involved? Uh, the test net launches five EVM chains. Uh, we're doing this as an MVP process, so we're steadily adding features to it as we go. But you can deploy today. We've got uh, and it's important to note here, this is vanilla EVM uh, with Pectra. So,

you can deploy today, right now, onto our chains. We have the hard hat plugin ready for you to go. Uh so there's nothing stopping you from lifting and shifting your application over to Cadenna. And uh due to our unique token economics, you know, even though we've been running since 2020, the foundation still has money. So we have a $50 million EVM grant program. So come talk to us. We we we have lots of problems we need to solve on this network. It's a great opportunity.

Um and of course, we're here. uh please come talk to us at our booth. This QR code takes us will take you to our uh testnet portal. Um from there you can apply from McGrat, join our developer communities, learn more about our technology, uh download the devkit and get started. Um we're really excited to be participating in the EVM Ethereum world and we can't wait to see what you all can do with it. Thanks a lot. we have time for a uh if we have time for questions I think we can take like one or two questions just a minute I mean this is a question that I always ask when I hear parallel chains so how

do you do atomic swaps between uh there's no need there's no need it's not based on atomic swap it's based on the original uh SPV that predates Litecoin so you just you just mint and burn uh sorry you just burn and mint it just shows up on the other chain. Okay. Yep. So, you said because the chains communicate with each other, there's no liquidity um fragmentation. Uh but uh how does that look like for a let's say a unis swap pool uh to have some of um balance of some token on all the chains?

Do they have to deploy on each of the chains with the same address? There's no need to so there's no need to deploy on all chains. you deploy on as many chains as your application throughput demands. Um so and you can consider uh different usage uh usage approaches. So for instance you could um you know you could have an alt with various stables on a given chain or something like that. And then there are uh there are ways to make liquidity appear instantaneously on other chains. Um but one of the ways is really it would be custodial something like I mean one of the great things about this is that intents on our network function really well because they clear in like a minute so it you know so while it's not like

instantaneous instantaneous necessarily for certain operations um you can rely on the you know you can have an intent and you can you know trade across our various chains but it it settles like that so it's a very unique experience Thank you for the question. Any other questions? You have any question? No more questions. Thank you very much for speaker. Thanks again. [Applause] [Music]