Filecoin Looks to Move Upstack From Just Archival of Enterprise Data to Database-Like Functionality

Audio-Only Podcast

Editor's note: Find the Blockchain Journal Podcast wherever you listen - Subscribe to our content on Apple Podcasts, YouTube Music, Spotify, SoundCloud, and Pandora.                                  

  

           

How Can Blockchain Take the Place of a Database?

David Berlind: I am David Berlind with Blockchain Journal. I'm coming to you from Davos, Switzerland, where [the] World Economic Forum is currently taking place. And standing with me right now as Stefaan Vervaet. Stefaan is the head of growth for Protocol Labs. Protocol Labs is the research arm that's behind Filecoin, which is one of the leading places that you would store information and data on blockchain. And you just finished giving a presentation on storage on Blockchain in the enterprise. Stefaan, thanks for joining me on the show.

Stefaan Vervaet: Thank you. Thanks for inviting me.

Berlind: It's great to have you. So, the reason that we're roaming around World Economic Forum here is to find out about how blockchain is being embraced by enterprises. But before we start with that, could you please tell me what exactly it is that Filecoin does?

Vervaet: Good question. So, Filecoin is the largest decentralized storage network out there. Our goal is to store humanity's most important information. So, what we've seen is that there's a need for a new way of storing data that is allowing customers to trace, verify, and protect their own assets. So essentially, what we're doing is we're putting the control back in the user's hands by building out this fully decentralized, blockchain-enabled storage network that allows any enterprise or end consumer to store data on this network that is 95% cheaper than what you would pay for at Amazon, Google, Microsoft at this point is fully mutable, meaning no single entity has full control over the network because it's decentralized, which is one of the benefits obviously of Blockchain so that one, customers can store the data in a cheaper, more cost-efficient way, but also the data is immutable, which means you can't change the data, and if you would change it, it would be immediately identified.

And then three, it's verifiable, meaning the data gets verified on a daily basis to ensure that your data is still intact and is still exactly the same as you originally stored it. So these characteristics of Filecoin make it very interesting for companies like UC Berkeley to store their data assets on Filecoin because it's a great alternative to public cloud or on-premise solutions that may cost a lot more, or customers are looking at storing data in certain geographical locations where there is no Amazon or there is no Google. That's where Falcon can offer a solution.

Berlind: When enterprises usually store all their data on-premises with their own databases, Oracle databases, SQL databases, Amazon came along and introduced the idea of cloud storage as storage. It took a long time for them to get used to that idea, but now cloud storage is a big thing. You are talking about maybe the next revolution in storage, but you also talked about immutability. And I think that when you look back at how enterprises use something like a database, they actually do appreciate the idea that they can go in and edit a record. So when you talk about the immutability of the data that's stored on a Blockchain, which is one of the big characteristics of anything on blockchain to what use cases, is that immutability suited because once it's on the chain, you really can't change it. And that's something that enterprises actually do a lot of.

Vervaet: Yeah, that's right. So, one, we are a platform. So as a platform, our goal is to create the tooling and the capabilities so that customers and software providers can build a middle layer between the databases that are out there today and the distributed Blockchain that we provide. And so the immutability is crucial definitely when you're looking at archives. So that's why we're seeing our initial use cases are typically customers that are moving data from tape or public cloud or on-premises, and they have these long-term data preservation requirements, like three years to 10 years even, where data every single bit matters. And so, for them, the immutability and the verifiability is super crucial. Second, the database you were describing is the next phase is where there's new technologies being built right now, new adapts, [and] distributed applications that are building the equivalent of what we know as a distributed database today, but then on top of Filecoin. So we're doing it in a staged approach.

Our first step is [to] go after the archives, and second, is then improving our retrievability and increase the performance so we can deliver better performance and faster performance to lower latency applications. Today we're not a perfect fit yet for databases. We're fit for archives and backups. And then, over time, as we improve our performance and our retrievability, we'll move up the stack and get closer to more like the real-time applications over time.

Berlind: I was inside the presentation, and I was watching the gentleman before you, Jonathan Dotan from Starling Labs, and he was talking about the preservation of war crime data. So that seems like one of those things that's really important because that data will have a role in the ongoing prosecution of war criminals and things like that. So is that a really good example of where that –

Vervaet: Totally. That's an amazing example because they capture the data at the camera, at the actual endpoint where data's being captured, and they track the hash and the footprint essentially of that video footage all the way from the edge, all the way to the background storage. So this is a perfect example of a fully integrated end-to-end workflow where they used the hashing technology to identify the integrity and actually the actual truth, the fact that there was someone on site, who identified there was a war crime, it's all logged, and it's traceable all the way back to the backend storage of Filecoin. That's a perfect example.

Berlind: In your presentation, which I sat through, which is about storage on Blockchain in the enterprise, you talked about the problems that enterprises have with egress costs. So let's first talk about what an egress cost is and why that's problematic for enterprises.

Vervaet: So we've talked to a bunch of customers. So one, we did research with IDC.We reached out to more than 300 enterprises and asked them," Hey, do you know about these new decentralized storage technologies?" And [the] majority of the enterprise actually have heard of it, which is great. When we asked them what their major bottlenecks were today with existing cloud providers, egress,41% of the enterprises claim that egress costs were so expensive.

Berlind: That's the cost of getting off the platform.

Vervaet: Exactly. And so what it actually is, it's actually the cost of it's cheap to move data into the cloud, but it's extremely expensive to take it out. It can cost a couple of hundred thousand dollars to take out a couple of petabytes out of the cloud. In certain scenarios, it's almost as much as setting up a completely new environment. But unfortunately, you need your data. Those are your most important assets. And so what we're doing with Filecoin is because it's an open marketplace, there are no fixed Egress costs. So it's really up to the ecosystem to the community to decide what those prices are and be –

Berlind: How does that work? When I'm working with cloud storage, let's say Amazon's S3 or even on-prem storage from some company that provides that Veritas, a company that I believe that you're used to work with, how is it that somebody else is setting the pricing for the cost of putting data on-chain or taking it off?

Vervaet: It's a supply and demand. So, we have more than 4,000 systems in our ecosystem. And when you look at our network –

Berlind: These are all the nodes that are on the chain running the chain?

Vervaet: These are all the nodes, exactly. And those are hundreds or close to a thousand entities, companies that are actually running these nodes. And, some of them are more focused on wholesale, very cost-efficient storage capacity. Others are more like white gloves, [they] provide a white glove service. They will come into an enterprise customer, they will provide storage services. They will capture the data and crypto data store it on the network that will create connectors. My point in this is that depending on what you're looking for, you will be able to shop and you are able to shop around today and talk to different storage providers and negotiate your own contract based on whatever your needs are. Sometimes it is, you need a certain, let's say, a storage provider in Belgium or in San Diego, close to where the research is being done because you need fast performance, or you need the data to be stored within certain country boundaries because of GDPR or certain regulation requirements. So that's all feasible.

And so in our network right now, there's an oversupply, meaning there's 15 exabyte[s] of capacity sitting out there, and there's 500 petabytes of data actively stored today, which is amazing. So, every day, we add around four petabytes of data onto the network. And because there's an oversupply, it's also very cost-efficient because a lot of storage providers are looking for clients. And so right now, we're in this position where the cost of storing data on Filecoin is 95% cheaper than an Amazon or Google market.

Berlind: But that cost could change as the demand for the storage capacity on the network goes up, and there's less of it around for anybody else who's coming on. Is that where the price might change?

Vervaet: Absolutely. So we do expect that the market will automatically balance itself out, meaning the price will balance itself out. We still think that's going to take a couple of years just because of the network still growing on a daily basis, plus the use cases are growing as well. So this is not a... It's very hard. You can't just look at cost by itself because there's huge benefits of using a decentralized network, as you know. Today we store, for example, more than 120 million NFTs from OpenSea and so on. And so it's our belief that every digital footprint that you create, all data sets that you create, should be stored as an NFT. And so that is not something that's feasible, it's not something that cloud providers can offer today.

And so there's a lot of new functionality like data that are being built that is going to be only available on a decentralized stack. And so my point with this is that over time, as the pricing will stabilize, also more functionality will come available in our network that is differentiating, that customers are willing to pay for, or customers are looking for, that will not be available in what we know as today, the public cloud providers.

Berlind: It's early days, and you're building a platform here, and the platform's just going to have more capabilities for enterprises to tap into. We were talking about the egress costs and the cost of getting off, but what you didn't talk about in your presentation are the ingress costs. And I think one of the interesting things about ingress on anything Blockchain is, whereas with somebody like Amazon or a traditional storage provider, there's a contract you have to sign, or you have to drop a credit card ahead of time, and they bill you on a monthly basis. Blockchain is more of a coin-operated scenario. You don't pay until you put the very first item on the chain, and you only pay for that little bit. So talk a little bit about the enterprise reception to that business model and the ingress costs.

Vervaet: I think it is a great question. So, how data is moving into the network, as you said, we're moving to a different model because we're using a wallet, we're paying in tokens. The way the place we're at today is where storage providers that are typically not only providing the capacity of the storage are also the ones that are also collaborating with the end customer and are helping data to be onboarded. So they're the ones that are actually going out to the end customer, are encrypting the data, transferring it either physically or over the wire, meaning offline or online. And so they're the ones right now that are obfuscating this whole notion of tokens. So they're closing an agreement with the end customer in fiat at this point, and they're using middleware to automate some of that.

Berlind: It's like the old value added reseller model where somebody builds up a solution on top of your platform, they sell that solution to the [inaudible 00:12:26]. Because one of the barriers to adoption for enterprises is just the idea of having to hold the Cryptocurrency that's necessary to pay the fees that make it possible for them to pay in Fiat currency. Right?

Vervaet: Totally. So today we see our source providers as a channel. Exactly. And they're helping those end customers move on-chain without telling them or sharing... No, of course, they'll tell them. But without sharing them all the complexity, I think over time in the next couple of years, our goal is to get enterprise customers also more comfortable with having tokens on their balance sheets and so on. Obviously, that's a longer road that we're taking, and that's why we're collaborating with Ernst and Young, EMD, and Seagate. And, we have a ton of other partnerships in the works to help build the processes that these enterprises will have to go through to adopt some of these newer technologies. And so today what you're seeing is a similar model as 16 years ago where the cloud providers came out with S3 and other cloud services is like these gateways that are office skating.

Back in the day they were filed to S3 or filed-to-object gateways that were simplifying the complexity and how data was moved into the cloud in the same way this is happening today. So we're building these on-ramps or these gateways that simplify for enterprise customers, how data is being moved into this network. So we're seeing S3 gateways today that are being built on top of Filecoin so that customers can just point their applications instead of, like in Amazon to Filecoin, but through these gateways. And then over time, what we'll see is more of the applications in the enterprise space will natively talk to Filecoin over time. This will take a couple of years. And that's also why in Protocol Labs, we hugely invest in startups. So we have more than 500 companies today that are funded and are building on top of Filecoin. And it ranges from consumer products that are competing against Spotify. So we see Spotify alternatives, Zoom alternatives, all sorts of platforms that exist today in Web3, but are now being built on top of Web3 networks and in infrastructure. And Filecoin is one of the key components there because every application needs to store data. And that's really why we are investing so heavily in this Web3 movement. And at the same time build the bridges, the connectors with the Web3 existing Web2 enterprises.

Berlind: Sometimes I call that Web 2.5.

Vervaet: Web 2.5. It's really better to say Web 2.5. Yes, because there is a lot of Web2, but it's really trying to make it transparent for enterprise customers to take advantage of these new technologies. And then what will happen is eventually some of these Web3 startups will eventually overtake some of these existing Web2 applications. And we'll see some unicorns come out of this. And we're very excited. It's a very interesting world.

Berlind: Okay. So now you guys have taken over this really beautiful old church here on the main promenade going down to the Congress Center. You're steps away from the Congress Center, where the World Economic Forum is taking place. So you're probably getting a lot of traffic coming out of the Congress Center. How's your World Economic Forum been?

Vervaet: Oh, for me, it's my first year.

Berlind: Same by the way.

Vervaet: Okay. Well, I'm impressed. I'm actually overwhelmed. It's amazing. Definitely, here in the space that we have the people that are coming through. We had a whole delegation of [the] US Congress yesterday. We had the founder of [the] Worldwide Web did a talk with our founder of –

Berlind: Was Tim Berners-Lee here?

Vervaet: He was here yesterday. And so he did a talk with Juan, our CEO, who started IPFS and Filecoin. And it was great to see those two leaders, the one that built Worldwide Web and Juan, who is, because we're building a new infrastructure for decentralized web, so the new web basically. So it was great to see those two leaders together and see the old and the new come together, which is great.

Berlind: Yeah. And Tim Berners-Lee is a big fan of decentralization. He built out this new thing called The Solid Project, which is all about decentralization. So Stefaan Vervaet, thank you very much for joining us here on Blockchain Journal in Davos, Switzerland.

Vervaet: Awesome. Thank you for the interview. Thanks for coming.