Hedera Describes Criteria That Enterprises Should Consider When Investigating Blockchain Technology

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.                                  

  

           

What Are Some Enterprise Concerns When It Comes to Blockchain?

David Berlind: Hi, I'm David Berlind with BlockchainJournal.com. I'm coming to you from Davos, Switzerland, where right now the World Economic Forum is taking place. Lots of world leaders, business leaders gathered to the Congress Center down the road right from this studio. We're located right next to the main promenade and I've been going around the entire event looking for people who want to talk a little bit about blockchain. We've been looking for a lot of the layer one blockchains to talk to, and one of the ones that's here is Hedera. And in the spirit of full disclosure, Hedera is one of blockchainjournal.com's sponsors. They helped get us off the ground. And I'm speaking with Christian Hasker, who is the chief marketing officer of Hedera and Swirlds Lab. So, thank you very much for joining me here.

Christian Hasker: Great to be here.

Berlind: Yeah, so it's good to have you. And we've been going around talking to some of the other folks who do layer ones, and I wanted to make sure that, in fairness, I came back here and gave you a shot at talking about Hedera as a layer one.

Hasker: I appreciate that.

Berlind: And so there are a lot of layer ones out there as long as we're on that topic. If you went to Matsari, many and also a lot of smart contract enabled layer ones. Because, I think, in general for blockchainjournal.com, our focus is going to be on many of the Layer 1s that allow enterprises to customize and write their own code to run on top of those chains. And Hedera is one of those chains that you can build smart contracts on. So in the spirit of all of these different Layer 1s, they all have unique value propositions to separate target developers, audiences, and organizations. What is the primary target of Hedera?

Hasker: So, great question David. And Hedera really was conceived from the get-go to be a little bit different from other Layer 1 blockchains. So it is designed from the ground up to be highly, highly scalable with very fast finalities, so low latency and at a very cheap, predictable cost of each transaction on the network. So that's on the technology side of things. And then the other important thing for layer one protocols is the governance model. So again, from its inception, Hedera was designed to have very strong governance. What does that mean? It means that we are governed by a group of global organizations, large enterprises, and also universities. And that was to have a very high trust model. So our primary focus is to be the chain that is adopted, mass adopted by enterprises. And we believe that there are obstacles to adoption that we have very good solutions to each of those challenges.

Berlind: Okay. So, you mentioned a few criteria. And the reason I want to ask you these questions is because when I speak to the purveyors of other chains, I discovered that everybody has a different definition for what some of these criteria mean. For example, when you take the word scalability, some people have one idea of what scalability means, and other people have other ideas. So in the Hedera world, what is scalability?

Hasker: Yeah. So well, I think in any technology, and especially when you talk about enterprise software, there's scale up and there's scale out. And every blockchain network should be deficient, not deficient, sufficiently decentralized with enough nodes where scale-out is not a problem. What does scale out mean? It means that you have a ton of redundancy in the network, right? Nodes can go down and your applications still run. But when we think about applications that are enterprise-centric and really business-critical, and frankly we're seeing so far very few of those in the market, we're seeing a lot of proof of concepts.

But when you think about what an enterprise needs to deploy any technology, they think about scale in a way that maybe a startup doesn't. A startup is focused on gaining their first set of users, whereas an enterprise is like we have to support out of the gate, if we're productionizing this system, it has to be able to be resilient and support these massive workloads. So, what we mean by scalability is the ability to support a transaction volume that an enterprise needs at scale. So that isn't tens of transactions per second, that is not even maybe a hundreds of transactions per second that is thousands of transactions per second.

Berlind: Well, also, though, it's a public ledger if I'm not mistaken, right?

Hasker: Yes.

Berlind: And so it's not like when you buy an Oracle cluster, and you set the thing up, and it can support some number of transactions per second across that cluster. It's a shared network. So, it's not just about one organization being able to reach its transactions per second. It's about the combination of organizations that are sharing that ledger and the entire load that they're putting onto it and making sure that it can sustain that entire load so that no individual user of the ledger experiences something like an outage. Is that correct?

Hasker: Yes. So you are competing for shared resources, and there needs to be ample capacity in your overall network to ensure that whoever is participating in the network on those shared resources, you can support that application workload. With Hedera out of the gate in a single shard, we are able to support throttled today 10,000 transactions per second. Those transactions are the type of transaction that would be me sending you cryptocurrency. When you talk about compute-intensive smart contract-type transactions, we can do maybe 450-500 transactions per second of those.

Now of course, if you have 500, to your point about shared resources, if you have all of your smart contracts using up that capacity, then you can no longer do your 10,000 transactions per second on your simple cryptocurrency transfer. So yes, it is definitely what blend of transactions you want to support. But as we see the volume scale, we can do two things. One, we can increase the throttles very easily and scale way beyond 10,000 transactions per second. And then the other thing that we can do is go to a sharded solution. And Hedera has some very, I don't know how deep you want to get into this, but Hedera has some very unique aspects when we even think about the future of sharding and the impact on security of the overall system.

Berlind: Yeah. For our audience, who probably knows very little about blockchain at this moment, what do you mean by shard? Because that word does come up when you're talking to the different providers of the different chains.

Hasker: Yeah. So you can think of a shard as a set of computers that is processing transactions and coming to consensus on the order of transactions. And when you use up the capacity of that network, guess what? You need to stand up another network with another group of computers, coming to consensus on those transactions as well. But you need to make sure that each shard in and of itself is highly secure. And as you add shards the overall system itself remains secure. And we know that we have very good solutions there as well.

Berlind: So right now, I'm assuming because you're on a single shard, you haven't really broken a sweat yet. You haven't reached capacity, and you're supporting –

Hasker: Nowhere one near it.

Berlind: Nowhere near it. And so you're supporting the loads that are on it, but at some point if you start to reach that capacity and you'll see that coming, probably, you'll be able to prepare another shard. What about finality? You mentioned that as another term and in my studies over the last six months, six to 12 months, I've discovered that finality means different things to different people. What does finality mean to you guys here at Hedera?

Hasker: Well, that's an interesting comment because, actually, finality shouldn't mean different things to different people.

Berlind: Well, marketers, if they don't do well on something, they will stretch the definition.

Hasker: Oh, yeah. Okay. Okay.

Berlind: So that's what I've been bumping into.

Hasker: Okay. So, let's talk about non-blockchain systems for your audience for a second. When you go into a Starbucks and you swipe your Visa card because you bought a latte for $5, there is a moment in time where your transaction, your dollar transaction, is committed to the Visa network. They know that you can afford to pay that, and the transaction is finalized. It has gone through. In a blockchain system, it is the same definition. At what point do you know for sure that enough of the network has come to an agreement that your transaction is now final?

Berlind: And cannot be walked back?

Hasker: Exactly. Cannot be, yeah, cannot be revoked. And the interesting thing is in more of the traditional blockchain technology-based public networks, they have something called probabilistic finality. And what that means is there is never a moment in time that you know for 100% certainty that your transaction really is final. But there is a point in time where enough of the nodes have come to consensus, and you say, okay, that's good enough. And that works fairly well.

Berlind: It's sort of establishing a confidence level.

Hasker: Exactly.

Berlind: It's not a 100% confident.

Hasker: It's not a 100% confidence.

Berlind: But it's so confident that we're pretty sure we can move forward.

Hasker: And that right there is the difference with Hedera. There is a moment in time on Hedera with the probability of one, which is a 100% probability that all the nodes in the network have seen your transaction come to [a] consensus, and it is now been finalized, and it cannot be revoked.

Berlind: Okay. What else should enterprises be looking for when they start to look at blockchain as a technology, as a service, or as some use case?

Hasker: Yeah, so the other thing I mentioned just in passing is security. Right. So enterprises, they think about throughput, obviously, can this thing that I'm choosing, can it support my application workload, but also how resilient is it? We live in a world where there are distributed denial of service attacks all the time on public clouds. AWS, a region, can go down from time to time. So, I don't want to get too in the weeds on what these terms mean. But in Hedera, we have something called Asynchronous Byzantine Fault Tolerance, which is the highest degree of security that you can get in a distributed system. It means that we are resilient to distributed denial of service attacks. And even if the internet itself is compromised, as long as some messages can make their way to the network, the network will come to [a] consensus on those messages. So it can function in the face of slightly below one third of the nodes being dishonest. No network, once you hit one third of nodes being dishonest, no network functions.

Berlind: No layer ones –

Hasker: No layer one. That's as good as you can get. And then the other aspect is that we just assume bad actor[s] exist and are out to get you, and things can be compromised, and you're going to have civil attacks, and you're going to have DDoS attacks. And our founder, Dr. Leemon Baird, what he invented was to solve for the throughput problem with the highest amount of security. So that is a trade-off that you have to see other chains making all the time. They have to choose: am I going to be fast and less secure, or am I going to be slow and more secure?

Berlind: It's interesting that you describe that as security because it sounds to me [that] you used the word resilient after. It sounds like, to me, more like reliability. That despite whatever gets thrown at the ledger, it's going to continue to operate in these kinds of difficult conditions. And I would say reliability. When you think about how enterprises look at the different technologies they buy, there are some criteria from one technology. It could be blockchain, databases, or application services[s]; it doesn't matter. There are some things that are just critical that apply to all of them. Security, of course, is one of those, reliability, [and] total cost of ownership.

Hasker: And I think reliability is a great term because it applies both to the throughput, right?

Berlind: Yeah.

Hasker: Am I going to hit a max number of transactions, and my application stops working? That's reliability. And does the things stay up all the time, even in the face of attacks?

Berlind: I just think about in recent history, when cloud first came up, and Netflix was on, and Amazon and the Amazon cloud went down, and Netflix went down there. I'm sure there was an uncomfortable call that took place between the CEO of Netflix and Jeff Bezos or somebody who reported to him.

Hasker: Yeah.

Berlind: Because so as long, that's an enterprise app, and as long as they're down, they're losing money and maybe losing users. Right. So.

Hasker: In another interview at another date, we can talk about that. I worked with Apache Cassandra. Adrian Cockcroft was the chief architect of that Netflix system that deployed the first massively scalable database in Amazon Web Services. And yes, you're absolutely right. They were very focused on reliability.

Berlind: And so that's going to be a major criteria item for any enterprise that's picking a chain. Because when they're going to pick something, I think they're going to pick something strategic.

Hasker: Yeah.

Berlind: You don't have [a] chain for this and one chain for that. You want to pick something that you –

Hasker: Yeah. You talked about [the] total cost of ownership as well. So, this is incredibly important to enterprises. Enterprises have to budget and they have to budget predictably. And so that's hard in the blockchain world because you have to pay for your API calls to your network. Whatever network you're using, when you're paying for your resources, you are using the cryptocurrency of that network to pay for the resources.

Berlind: You can't use fiat currency. You can't just pay in dollars the way I pay for Amazon?

Hasker: I mean, there could be something built on top that abstracts that away, and you pay a vendor, and they pay, the API calls on your behalf, but the actual API calls itself to the network are based in the native cryptocurrency. And that's a challenge if you have your pricing in your cryptocurrency because, as we've all seen, even if you don't know blockchain, you know that cryptocurrency is a volatile market.

Berlind: When you say you have your pricing in your cryptocurrency, you mean that it costs us this much amount of cryptocurrency to do that transaction?

Hasker: Yeah. So let's say we are a network, the David Berlind's Network and we have David's right, and we're paying in David's. So you may say this resource costs one David.

Berlind: I just retired. By the way, there's a lot of David's out there. Go ahead.

Hasker: This resource costs two Davids. This resource costs three Davids. But as an enterprise, you have dollars or sterling or Euros, whatever your currency is, and you have to budget for that, but you have to budget like okay, the cost today for me is one David is $1, but tomorrow one David is 10 cents, and three days later, one David is $10. How can I predictably budget for this system? And we are governed by this global organization. And one of the, there's a pricing committee who makes these decisions and one of the very first decisions at Hedera was the initial council members saying the pricing model has to be predictable. So, how do you do that when your currency is unpredictable? You price all your services in dollars. Yes, you pay the API calls in your cryptocurrency, but the pricing is done in dollars. And that means that for us, a simple cryptocurrency transaction is one 100th of a penny, always.

Berlind: No matter what.

Hasker: No matter what. And then a smart contract call might be a dollar depending on the amount of resources you're using, but it will always be a dollar.

Berlind: So, does that mean on the back end, you're essentially sliding the cost in your cryptocurrency? So –

Hasker: Correct.

Berlind: If –

Hasker: Yeah, every 10 minutes it gets –

Berlind: Okay. I get it. So the actual cost in cryptocurrency is changing how much –

Hasker: It's the exact opposite of what everyone else is.

Berlind: Yes. Okay.

Hasker: Yeah. But enterprises love it. And it's frankly hard for me to, having been in the enterprise software space for many, many years, it's just hard for me to conceive of some of these procurement teams. They're not going to –

Berlind: Well, that's got to be a non-starter because I've been doing this kind of work for 30 years. And every now and then, you have to make a proposition to somebody and your boss. And then the question is, well, how much is it going to cost over a period of year? And if you can't do the math because there's some volatility in the math, then they're going to send you away and tell you to come back when you have the answer.

Hasker: That's right. Yeah.

Berlind: So yeah, I can totally understand that. Well, Christian Hasker, Chief Marketing Officer for Hedera, it was great to sit here and meet with you.

Hasker: Yeah, thank you so much.

Berlind: Yeah.

Hasker: Congratulations on Blockchain Journal.

Berlind: Thank you so much.

Hasker: By the way, I know you've had some great interviews here in Davos. I haven't watched them all, all the way through, but I intend to catch up.

Berlind: I greatly appreciate the support and look forward to keeping an eye on how things go. And good luck with the rest of your World Economic Forum.

Hasker: Thank you.

Berlind: Yeah. Okay.

Editorial Disclosure: This article mentions the Hedera public distributed ledger. At the time this article was published, Swirlds Labs, Inc. sponsored the operation of Blockchain Journal and is also part of a multi-party coalition that contributes to the success of the Hedera ecosystem. For more information about Blockchain Journal's policies regarding editorial independence and neutrality, please read our welcome to Blockchain Journal.