BNB’s Decentralized Storage Network by Arno Bauer

foreign I'm the co-founder of the Norway and also CEO actually in Norway is one of the key contributors of the BMB chain including this BAC and zkbmb and recently we also announced the Optimus roll up for the BMV train absolutely we are super active for the Springfield contribution as well we are very happy to share this latest update for the Green Field and over to you yeah sure thanks Ben so first we're going to discuss about like what is your central storage and what we have today or previously and why we want to move to the decentralized network as well so previously as we know like Enterprise invests a lot of money inside storage inside cloud storage as well so they have huge data set but everything is more centralized nowadays so that is like how we can move from a centralized storage into a decentralized way by helping Enterprise as well to move to the new storage business there is really like how we can provide programmability on storage as well in a digitalized Manner and how we can provide Finance visualization as well with data something that is missing a bit nowaday in the recent price environments like we centralize storage people can already monetize data but they don't really own it so the aim is really like how we can move to the next stage so one of the main innovation of Greenfield as well and let's maybe roll back a bit on why we created this white paper as well so as we've been seeing in the past three years we had the rise of D5 with the decentralized finance platform then we had nft digital ownership and now it's like how we can create something new that can really help users to truly own their assets and monetize on top of it so this is basically the aim of this new white paper which is BNB Green Feed like how can people truly own the data and monetize it so one of the main innovation of Greenfield is really rooted in this deep connection with BNB smart chain basically BNB smarter is our evm compatible blockchain and Greenfield is another Ledger but we basically have key connection between both which is really going to help us to monetize and program on top of this data so we have as well specific base tool which is like the account is going to be shared between the BSC and BNB so for a user is going to be truly seamless to interact with both blockchain and as I said before the user is really going to be able to truly own the assets more freely and independently as well and monetize on top of it so let's imagine you are like I'm not sure you have a Blog you want to publish your work on a publishing websites so how you can really own the data that you published and monetize it as well resell it so this is what Greenfield is truly made for to truly own your data and automatically resell it in a programmable way using smart contracts I don't know if you want to add something on top of it Ben or maybe later yeah okay sounds good uh so here we can see like a little diagram of how the cross chain is working between BSC and Greenfield so it's really something that the BNB team as well is used to do because we've already have a dual blockchain system so now we basically extend it to assert layers so basically we have on this side we have Green Field network with the validator the storage provider so it has his own Ledger which is tender Man based and then on the other side we have the Smart Train which has the application layer the resource monitoring and the cross chain contract which is building as well so as we can see we have different layers which are basically going to handle the forwarding the broadcasting of events from one chain to the other and in this way for example like a smart contract is going to be able to change the ownership or change the read access to a specific file that is called on Greenfield of course Greenfield is not going to store the data itself The Ledger is going to store the metadata of this file but the file is going to be stored on a storage provider so I'm going to talk a bit later about the storage provider part and how it is playing a role as well in Green Fields so here we can see the core like what's under the hood so basically Green Field it's a dual system right so you have Greenfield Ledger which is Cosmos app based that's the validator very data are going to do proof of challenge towards the service provider and on the other side we have a network of storage provider so we have multiple storage providers that are basically providing storage to users and which are basically High storage service infrastructure with high redundancy and data availability as well so basically Green Field validator are going to challenge as well storage provider to see if the file is available or not and based on the challenge if the storage provider is not able to comply and show the proof of data then they are going to get slashed basically to be asteroids provider you need to stake a total collateral in order to be active in the network what can I add as well on this one uh I think next slides may be more important to explain that the whole structure I believe uh yeah uh I will I will first send this one so just about the storage provider for the users so we are going to have multiple storage provider not only one of course so the user is going to select his primary storage provider and then we have secondary as well that you will see later on the different slide but something that is key in Greenfield is that format fee point of view the user is going to see the fee in USD and not necessarily like in a token or at crypto assets so for the user it's going to be something that is really like tangible like USD is something that they are aware of they can compare with other system as well but under the hood the system is going to do the payment flow using BNB token so which is something that is quite Innovative as well in the industry um so here I'm just going to describe a bit more about the validator and what they are doing what's the purpose of the validator as well so as I said before Greenfield by data is a cosmos SDK logic it's a tender main course so it's Byzantine for tolerance and the validator as well are going to be able to they need to stay collateral in order to be active as a validator on Green Fields one of the main mission of this validator is to prove the challenge to the storage provider to ensure that the file is here so one way for the validator to get reward will be transaction fee proof of challenge so if there's potential fine to stretch provider they're going to get the reward as well and then we are thinking as well of other reward mechanism that is still like in a white paper stage so we are looking forward for other people as well to help us in defining how the token economy might be sustainable in within Greenfield quickly about the storage provider as well so we can have multiple storage provider the storage provider is really going to be responsible to provide the access of the file the upload the read and the management as well to get clear read and download availability as well so here we can see as well the replica so how do we do for the data availability compared to others so unlike other decentralized storage Network green field data availability is incentivized from a financial point of view and it's protected as well is protected from a slashing point of view as well so we ensure that the storage provider is it's mandatory for him to have the file active because if he hasn't if he's deleting the file which could be here it's going to get slashed is bonded BNB for example so here we can see as well how the replica is working to ensure that basically there is erase your coding available so we see that we have the primary it's a storage provider on one side and then we have multiple other storage provider which are incentivized as well so they are going to receive a cut over the fee in order to save a certain bytes of the file as well to ensure that the replica of the file is like replicated among multiple search provider so even if the main stretch provider is going down all the other stretch providers still have a copy of this file as well and they are incentivized to have this file too so it's going to really push the competition within storage provider to provide the best service for the user here we can really see Greenfield in action so it's like a whole map of network it might be a bit complicated to look at it at first sight but on one side we have the BSC which is like the main evm layer that we have the smart chain on the other side we have Greenfield and we have like relayers which are basically going to be able to relay events when I say events like again change of ownership to read access right or add user to a specific file remove it delete a file and so on then on the bottom we have the storage provider which are going to store the file so the storage provider are going to send as well transaction into Green Field those transactions those transactions are going to be the metadata of the file it's going to be object based so like who has access what's the link how we can access this as well and I say the validator they have basically a watch mechanism so they need to challenge storage provider to ensure that the file is here as an application Level you can either you can directly interact with BSC and with the storage provider directly you don't need to go through Greenfield for example it's just basically the secure and the governance mechanism of the whole infrastructure yeah maybe I also can want to add a little bit color for this one especially for the developer right so just imagine when you are like doing uh you know upload or manage your like either image file or streaming like this kind of files right where do you want to start so typically you may start from this the actor upload this kind of information right but you need to make sure you know there's the by the way this this API right is like purely S3 compatible API right so this is something like a currently Green Field and no real we work together try to make sure this one compatible because we know a lot of mft projects they are using this API right now right using S3 however we want to provide it a very compatible apis as SS S3 right to make sure you know you can reuse most of your codes there and another important part is after you upload your data here right it will automatically you know create the Green Field resource and also Greenfield resource mirror so because the BSC is evm compatible which means you can go through this Green Field resource mirror I can access your storages right through this kind of access control as uh you know Arnold just mentioned so with that you can have a much much better resource-based programming right stuff to manage your like nft for your images for your like a streaming essential and of course later you may can see that how do they like manage this kind of access control for these parts so with that one we can link this storage and the evm together right through this kind of the connection there so that's the whole idea is to make it like a very seems like seamlessly integration between the storage and evm yeah I want to add a little bit color for this part yeah no exactly we really try to make it like inclusive for web 2 developers so it's is going to be easy for them to migrate to Greenfield because at the end it's not really a migration because apis are going to be standard like Cloud apis so they're really aware of like how to use it how they can manage it so from a business perspective it's going to be much easier as well for application to handle such kind of storage Network as well and again as I said the payment flow for the user it's going to be charged in the number is going to be a USD number so it's something that can really compare with other but under the hood everything will be in BNB so now we're going to show as well a bit of like some comparison like how do we compare Greenfield maybe compared to what's happening today and why as well we've made it so here we can see a nice comparison table that's not real team made like for us as well so we can really see like on one side we have green fields we try to compare it with like filecon our web or storage which are basically kind of the industry leader as well to the in the market so we can really select from a data redundancy perspective like we already like erase your coding by default so basically it's a bit like sort G as well compared to like a replica in other system which is really like ensuring a higher availability and guarantee of the file radio density so as you know like as a with the file I showed before like this erase your coding is really helping to make sure that even if one node or one storage provider goes down the file is still accessible and you can reconstruct it as well so it's something that is really important again for the data availability so we really have like unlike all the other we have a really like Financial incentivization for the data availability and as well as the punitive mechanism in case the file is not here anymore so we are really like punishing storage provider easier if they are not doing what they are supposed to do which means that over time the only storage provider which are going to be alive are the one which are really providing the best service for our users so there's going to be really a high competition among audio storage provider in order to get the most users does the most like Financial benefit as well at the end because they need to be profitable as well that's for sure you want to add something else uh sure so I think that you know if you read this kind of the whole diagram right this kind of comparison by the way I see somebody like takes a photo don't worry you just go to the no real you see this blog because we really published one blog named the three things matters for this to understand this Green Field right so there are three several major areas one is the how can we make sure the data are really distributed and also make sure it's safe efficiently right so through the through the different algorithms there so this data reductions and data availability persistent mechanism are for that purpose right we need to make sure these areas however if you go through this algorithm right so absolutely we are you know working with the bmh and try to refer to a lot of the best practices in from our FICO and storage right so you can see that we absolutely introduce a quite similar process there and at the same time right so the quite Innovative areas how do you protect who can access this data so this is maybe the quite a new Innovative idea for the Green Field right they really introduce native Access Control later I can show you know some of our because we for no real we provide the tooling right for this kind of enable Empower this one you can see some of our preview versions and another part is also you know we need to make sure for some of the data you do not need to always persistent like forever right you can have a temporary solution to to delete that so that's a majority for the data in the data access pipes and another part I just mentioned is for the programming and it you know consistent API right we want to make sure you can using your traditional evm understanding and your S3 access right this kind of API you can use that one right so um and the last part is the payment right so we need to make sure you have a very flexible Billing System right so here in the BMV Trend in the Green Field right they introduce some of the streaming payments which means pay by use right so you can very easily to use that one so with that maybe I can switch to the next page right to you know leverage this opportunity you know because we will nurio is one of the infrastructure provider so we want to provide a tooling to empower the storage adoption of this green field and also decentralized storage right so provide this kind of the UI if you are familiar with the solidity you may be or are we Etc you may be quite easy to understand that this kind of a look and feel right so basically you can select your primary you know they have a dashboard you absolutely can select which storage provider as a primary or like a secondary is based on your preferences right based on the ranking so we provide this tooling you can decide and then we have like a buckets right you can manage your uh this kind of the different files we have the similar Concepts as S3 buckets photo you can position into your assist into this green field and uh of course you have a very easy access control there uh through this kind of you either assigned to the you know group of the people through the by the way the group of people here means the group of the address right they have access to your fires so we have this kind of Access Control will be introduced here and of course you can have a oh sorry there's no payment um yeah it looks a one page I missed so we do have like a payments there as award is right you can support the payments and very easily or pay by use right so this is something like we are currently in the preview version I believe BMG maybe will announce the win that test native or Mainland right will be happening there right so we are still like early development so it's still white paper so there's a lot of stuff to still need to be shaped so that's why we want to have like your feedback as well nevertheless we are oops so if you want to have kind of an early access to the like the preview of the code as well so soon we are going to release our Alpha Beta as well so probably before end of this month so we are March right so yeah before end of March we should provide already some specific access to like early developers who want to help us to debug see what's working what's not working how we can enhance as well the documentation so it's really a community project as well so we really want to have the community which is helping us so maybe we have as well like some time for some q a and I guess there is a lot of questions probably from a peoples I know that some already discussed a lot so feel free to ask questions again thank you uh this is a great presentation I was wondering on the storage architecture slide where you all had the different um Storage storage users is that a hash of the concatenation of all the storage hashes or are you all doing some type of arithmetic operation on the hashes to get the um the actual identity can you repeat the last part I did not understand exactly yeah so on the on the slide where you all had the storage architecture layout right right to identify the storage providers yeah that one that's one yeah the last one yeah the storage segment sorry um that hash mechanism that you're doing at the bottom are you all adding the hashes together are you concatenating them and then taking the hash of that uh I'm not sure exactly yeah I think uh so far to be honest uh I think we work with the machine right now right is it still in the white paper but this is just like a initial idea absolutely you can just imagine right the data is still saving to the storage underlying storage provider but how can you link that way you have to absolutely you need to have this kind of hash to connect this one and of course this harsh need to link to if we go back to this right go back to the next page because this one the data is saved here but how can you make sure this data can be reflected into this Greenfield resource that's managed by the Greenfield blockchain and this one also be mapped through this layer between the Green Field blockchain and the BSC production which is a Greenfield resource mirror I think this hash will be linked together and you identify with the hash exactly so through this one we can have access control to this resource then you can access this file directly interesting okay that clarified so that's the idea yeah thank you very much sure I think hey uh sorry I was wondering how you guys are doing payment streaming so the payment streaming that's a really great question I know we are still working on it because it's one nice one of the hardest parts because we're going to have a so you're going to like if I'm correct we will need to look at the white paper exactly but basically you're going to have a kind of a credit and that credit is going to be consumed frequently based on a different type of agreement that you have with the storage provider so when you're going to upload a file you can as well specify how long you want this file to be uploaded is it one week two weeks three weeks one year and let's say you have done this credit of one year then your account is basically getting deducted on a I don't know daily or the basis I don't remember exactly anymore but then once you don't have this credit anymore the storage provider is it's his choice like it can either like keep the file for a bit longer like to give you time to come back basically or if it does if he wants you can basically delete the file as well so as long as the payment is not working anymore the storage provider can delete the file but it's going to be like as a stream of like every time there is an action basically the credit is going to get deducted from the accounts as we were working with the API right providing this kind of a tooling right now sorry we do not get a UI today but basically there are multiple type of payments right now right for example you you need to have a store which means you want to open the you know specific which quite similar is a job box way right you you need to pay how what what's the size for your storage so this is like a one-time it's not necessarily have like you know streaming payments however somebody will use your like this data that's a download right which which is can be like pay by use because they do not need to access your full storage so that's something like you know we're working with them for that one right now right they put not only like provide this kind of the you know mechanism but also SDK so other streaming payments can include with that but that's the idea we're doing right now yeah cool maybe in follow-up what was the reasoning beyond the decision to charge in USD is it are you expecting Enterprise or traditional kind of AWS customers so it's we feel like it's easier for a user to understand like what's the true cost and really how we can compare it with like the traditional storage providers like Dropbox Google Cloud everybody S3 so this is the it's the base currency that everyone is using when they want to use such kind of services so we want to make it seamless for the user because if we charge it with token there's too much valuation like an constant change right so you're are you hedging the USD to the the token price because if the token's going like this you're you're not necessarily charging like this every month are you so you need to make sure that you have enough BNB that's in your accounts to handle as well like the filtration I get it okay the payments you are going to make today might not be the same as you make in one month for sure okay but yet you still understand how much it will be like from a USB USD perspective I get it because at the end the user like or the builders when they're going to build the end payment from the user is probably going to be USD as well like when you build the application on top of it so I think it's really meant for Builders and not for like just retail user so let's say you are a publisher you want to publish a website where and a journalists can basically publish his artwork and then these journalists want to resell as well like a membership so this membership will be paid in USD for example so that's why it's more like build like that because it's really more for Builders and not really for retail thank you hi uh I have a question regarding this did an architecture uh from this data architecture I can say that this there is a resource uh mirror layer between PSA and Greenfield and use the reliables to synchronize the resource across this to blockchain will this be a latency concerns in the future because you have to keep the resource as synchronize synchronized into blockchain and there is always going to be efficiency cost in that especially when users create objects uh following another question that have you ever thinked about some other approach like I say other solution trying to do is that they don't they they don't use the synchronization ways to get the resource across to blockchain however what it does is that they keep the ship the capital resource separately created the modified just in one blockchain and use the JK row up to other channels which keep the efficiency also just want to know why you make that choice and what's the things behind that's a really deep question to be honest I think it's I mean I don't really know how to answer to be honest I think we will need to discuss more like with our core Tech Team and I did you scan the group as well to join the Early Access uh and my phone is out of power but I will let you after because I think it's definitely it's a great question and especially like when we're going to scale to a higher like to really be able to scale like this green villager as well or like when you have like a lot of file like for sure when you upload on your one file it's okay but if you upload like 10 000 file how is the network going to sustain itself exactly that's the key question as well exactly not only just about latency but also kind of use app or your transactions in the bssa because right now PSO say it's TPS like a 600 seven thousand so but with this architecture eventually there may be a lot of TPS will be occupied by the operation gonna happen in Greenfield and gonna be synchronized in in the pssa right but just make a difference So Greenfield is going to be like tender mint so definitely like the from my performance interview is going to be much higher the BSC okay but after as you select the connection with BSC how we can make sure that this one is scaling as well with the usage that's a good research to do I think through with our team to make sure that we're able to answer this question okay um so my question is like how exactly we we can use the Green Field like is there any rpcn point API endpoint or like any chain a bit louder okay closer yeah my question is like how actually can we use Greenfield like is there any RPC or API endpoint or like chain info that we can right so there's going to be like direct API endpoints like don't really like providing tools for it so maybe you can discuss like I don't know I do not get the question how we can use Greenfield like what are the tools that are going to be here oh when are you planning to deliver it yeah yeah I I think that you know there are multiple places right you can see that there's a different access layers to this one so actually if you see this one even without the PMB chain right BSC the green fader itself can be I can be exists themselves right so which means you can access the storage directly from here access storage from directly from here as well because the address we did not mention the account system the account system here are exact same means that you have a BST address you can you can access this you can have the other one right so here which means you through the evm all of this evm if you are the evm coder you can access all of the uh through the smart contract already you can access this one and if you are the data access uh you know operator let's say you want to get the download data or something right you can access if you have the access you can call this Stop Storage provider API which is exactly similar as S3 apis right very easily to get the data out so there are multiple access points right to get you connected I think for this one major they may be connected through the typical RPC way right to candidates and this one maybe you know goes through this very uh straightforward you know like a hash based way URL like based away right to access these files so that's the typical way to to access the you know this whole system yeah hopefully there isn't no Dramatical change for the existing program language and also user experience there so I asked this like before because like we wanted to use it um and like to make something that's why like um how exactly can we uh like is there any documentation that we can refer right so we are the core team is currently finalizing the alpha release so like you can join you can add me after and then I can add you as well too like and I can let you know once it's released Alpha should be out like in two or three weeks so it should be quite early and then we can stop like seeing like how we can exactly use it what is missing as well and like how we can what's the pain point that we might have in our future as well about scalability for example so I think in two or three weeks we might have like more clear information of like how we can connect how we can utilize it and how is the let's say flow from an application point of view actually all of this will be open source open Greenfield with open source right so which means absolutely you know uh welcome ball like Builders right to help us together to join this BMV training Community it's quite exciting we feel like you know we're building this kind of API and the initial Explorer we will build this Greenfield Explorer and also Greenfield uh like Joke Box application as the first contribution there if you have any like idea right feel free to to jump that I think even the talks maybe can help a lot all right so that's definitely we are looking for a lot of help like from the community to like increase and enhance the solution yeah I also have one more question like uh uh can you go to the slide where you compare other with other uh yeah um so uh how uh what exit what what kind of like datability proof like is uh is Greenfield using like as far as I know uh in Falcon they use uh like proof of space time to like um to to verify that they are storing actually they are actually storing it on the Chain like how is uh what is Greenfield using in there so for that we have the proof of challenge which is the one yeah my question was like what what what is that exactly why it's happening yeah so it's going to be the Greenfield validator so those very data are the one which are doing The Ledger basically and those validator one of the specific tasks that they have to do is to do this proof of challenge so something that maybe I forgot to mention so the Green Feed validator are completely separate from the storage provider so they are not you can have one entity which have one radiator and one storage provider but most of the time the radiator are not going to be the storage provider so they are running two different systems as well so of course like that's right the very data is not going to do the challenge towards himself like if this was the question it's definitely like the very nature there are different entities compared to the storage provider so of course like once juice but it does entity are doing the proof of challenge they have no let's say link with the storage provider so they cannot collide together for example yeah thank you thanks for everyone maybe I will just uh add this quickly so feel free to join us