RaaS and Oracles: How do they interact - Jakub Jaworski | Redstone

thank you very much hi guys uh I need your attention for next 20 minutes I know you probably after a long day but uh yeah pretty tired for sure uh in case of my topic uh in the previous lection there was a mentioned about oracles so I will explain a bit more about oracles itself but in the context of how they differentiate itself from others and also in the context of RAB as a services so yeah as I'm yakob I'm Gro lead At Redstone and let's start so uh to start the things off uh let me introduce you about Redstone so Redstone raised around like 8 million of fund um with the big FS like G base Ventures lamis cap MAV 11 so on we also did uh Angel rant with prominent Angels like sandp like uh am in from Avalanche or Alex lukovski from zikas SN and you can notice that uh all of them are uh prominent uh L1 and L2 Builders so they understand the need and importance of Oracle we also last year launch our research arm uh where we are focused about area that we are currently in ASU and uh this is like a very big uh overview about about market like lsdy like raking also stable coins uh yeld and earning products so I really advise you to like look into that if you are interested into def uh topics like that and we are lean team of around 15 people like mostly developers always in our Redstone vest on the event so you can easily recogniz us and uh on the other hand we are hiring new stuff so if you're interested to work in the top notch product uh please you are welcome uh also a bit of alpha here like uh we started uh with our final product in the first uh half of 2023 and for now we SEC we are securing like 3.7 billion of dollars of assets that's assets that rely on our price fits so uh let's start for the oracle itself what it is so in definition oracles delivers data to the deps for smart contracts execution so I like to say that uh oracles are connector between blockchains and real world data so it usually can like connect data like weather forecast or sports or betting whatever but in terms of crypto is mostly about market so the volume market cap but in term terms of crypto speculative and defi it is all about uh prices so what is the price of the Bitcoin was the price of The Ether and so on and actually like there is a this is like 95% of queries and there is a question for you guys what is the most utilized price fed in crypto any guesses you can shoot yeah it's if W is if slusd this is like a 90% of queries is based on that uh and why is that because uh of course obviously ethereum is the one of the biggest Network in terms of volume but also there is a growing number of Assets denominated in if so all assets like LRT LST they are like uh they have to be denominate into if USD so uh I will bring down the type of protocols that are using oracles so um right now oracles are mainly used in defi so type like Landing markets with a or Venus like cdb stable coins like liquidity and Prisma leverage protocols like gearbox and Delta Prime uh derivative protocols or perss like GMX or synthetics yeld protocols like SAR and enzym finance and algorithm St bit coins like menta and fra finance and question for you guys do you know any protocol that is not leveraging uh oracles excuse me Franklin coin I'm not familiar with that okay that's that's that's something that yeah yeah it is what else like this is very like big one yeah correct like dexes dexes are not utilizing or Oracles in fact the situation is reverted oracles are using like uh dexes to Leverage The like the data sources from it so why does it matter for RO up as a service like how to start so in terms of Ras right now the threshold to create a new ecosystem is like you can now do it in the few days like in two days you can have a new ecosystem new network uh because of the RAS so also if you have if you want to have like users on your network you should have uh Defi and this is connected with Oracle and what you can do you can go to Chan link and ask guys about hey I just spin up a new ecosystem yesterday and could you like come come and give me a price price feits and they probably will say like hey we can Circle back in the months or years when you're going to have like a more interesting stuff on your chain and um probably this is not the best solution for you don't get me wrong like chaning did a very good job in terms of uh creating defi because it was like a cost of catalyst for all defi space right now but uh do you know how many chains does chain link support right now that's a question for you guys you can you can shoot close very close so yeah to to to like to say it's last time I checked it was 13 so not a lot and uh to give you some comparation in our case is around 50 uh more than 50 chains uh of course there are some active and some not very active but uh yeah 13 is not a big number so why isn't chaning on all of these chains so K bottlenecks are onchain data storage cost and monitoring infrastructure they can consideration here uh update condition and cost of one update okay implementation push and pull oracles and I will explain you about those models so uh push Oracle introduced by Channel link uh in 2019 2020 uh well is based on set up a Rel layers infrastructure and then based of some update conditions like time interval uh or deriva deviation uh and based on this condition the price is constantly pushing onchain comparing to the pul Oracle when there is no per se upd update condition so the trick here is this is their data is pushing to the onch only when it's needed and is usually triggered by the user transaction so how how does it look it looks like that the data is stored in the uh offchain component called middle layer in our case this is we call it like data distribution layer which is like a stream of offchain nodes and then where is a trigger uh in uh there is a trigger like user is for example deposit into the Vault or there is a liquidation the price feed is injected into user transaction is and pushing on chain so push to sum it to sum it up um the the user is paying gas cost for pushing it onchain uh in terms of push Oracle this is uh not the case because you are paying constantly for those updates and believe me it can be like from $15 up to150 million of dollars in terms of the big spikes on gas uh gas on etherum Manet for and uh going further uh about cross chain so uh it's obvious that push Oracle is not very scalable because of this that you have to uh launch the Oracle uh on the each next chain and also each next uh price fed in terms of pool Oracle uh like from the Oracle perspective they don't care about which chain is going to be uh and number of fits uh this is very similar case because of uh coming back to the CR one more time uh about pool here uh like in terms of redstone we developed like a special only in one solution for all evm chains using cold data in the ethereum transaction I will elaborate on that later so uh here is a model of standard P push Oracle so there is a data sources mostly taxes dexes or uh stuff like coin gigo then data providers are collecting uh data from the data sources aggregate them and pushing to uh to the Target chains and from then D smart contracts are reading from the blockchain the price fits and in this model this is not very scalable across chains and number of assets so we like the comparation about push oracles with with an email because uh we believe that push Oracle going to stay with us and this is like current email because we are still using it but there is other more better solution like telegram like WhatsApp and so on uh they're going to stay with us but there's a better and more efficient way so uh to give you comparation how structure of pool Oracle looks like um the beginning is the same uh you have data sources and data providers but then data is signed and data packages are signed and sent to the middle layer in our case data distribution layer but for example it's also um in case of piff this is piffet which is a fork of Solana um and they have like a 100% supply of the coins so this is pretty cheap and then uh it the price fit is injected into user transaction when there is a trigger in terms terms of like like I said like maybe a liquidation event and uh there is a couple of ways like uh Bridge uh Bridge way which is changing Bridge piff is using so for the bridge messaging next solution is Liquidator and this is what we are using which is a piggyback user transaction and this is like a cold data what Redstone is using on that later so you can use you can uh go to this website and see how it looks like uh and test this model so what is the call data so call data is read only by addressable space uh and here like ethereum is a state machine so so here is a argument of the function the transaction and here we uh uh pass the price feed of the data so why does Russ should care about it uh when they spin up new ecosystem first um waiting time uh that's this is the thing that rats should consider when they want to integrate the Oracle SO waiting time pull Oracle is better so point for that because uh we can strictly at the beginning think of the uh R straight away integrated with you in terms of push this is there there's different because even if you say this is evm evm are different like uh there's some differen sh like uh signatures uh some infrastructure nuances uh and so on cost in terms of cost pool oracles are better because uh of the gas cost uh and types of applications you want to bring this is going to the push Oracle like why is that because most of the big defi applications uh are used to use push model uh and they like don't want to change the whole infrastructure interface because of the uh changing the model so for now push is winning uh but we believe in future this is topic to change so uh case studies of the rollas service so I think the biggest success so far is Monta Pacific using Caldera on the polygon Z zik VM of course for now like the liquidity is going from this chain but uh I think this is good success story also we have very old chains that are trying to get the user B uh back like Lisk uh which is on the gelato on op stack uh maybe we're going to see many more uh examples like that also for example cell with zkm the same Kanto and maybe we're going to see like I said more dino coins that will uh do more uh this kind of stuff uh okay so to to wrapping it up uh oracle's parameters to keep in mind when you are choosing uh Oracle as a rollup Uh custom customizability can I decide on which data providers to trust uh in terms of chain link uh you take it for granted what is uh they bringing uh redundancy what has to fail for the Oracle to go down so is there a component that is uh easy to go down like Bridges maybe uh if the bridge is going to go off if or does oracle going to uh work so on uh l2s and l2s and l1's Ras compatibility should it be cross chain number and types of data feeds available how easy is it to add a new one uh next one onchain versus offchain price sources this is very interesting because uh we can see the very growing number of assets that are only available on chain like LST or RT assets right now and price aggregation methodology is it robust against the wash trading it is uh there is a couple of methodologies like lwop twap what is the best for current solution historical mispricing and manipulation has it failed in the past and how guarantee do I receive damage payout pricing model and integration steps when and how much I pay governance and toomics um um fundamentals and probably 100 more aspects so that was it uh if you have any question you can ask now or after the presentation uh thank you for your attendance do oops do we have any questions okay I guess it great thank you very much thank you guys