 Hi, I'm Stu Miniman here at the SiliconANGLE Media Office in Palo Alto. Happy to welcome back to the program, Kiron Bagishpour, who's the CEO of Igneous Systems. Kiron, great to see you. Great to see you again, Stu. All right, so we've been really busy at the Cube looking at, you know, so many big trends. And of course, really looking at kind of, you know, massively scalable, distributed type of architectures or something we've been looking at and something that I know Igneous has been doing since the earliest days. But the exact focus of what you've been working on, I think, has changed a little bit since, you know, you first came out of stealth and we've been looking at what you're doing. So why don't you bring our audience up to speed? We'd love to do that. It's not changed so much as expanded, if you will. We launched, I believe I was here last, you know, October of last year, just as we were getting ready to launch. And at that time, we launched the company and the platform, which the beginning service was object as a service, delivered as a service into the Enterprise Data Center. And that was just the beginning. We've gone on since then, expanded the number of native services available. But really what we have done is built applications on top of that. So the first application that we have developed and deployed at customers is backup and archive for massive file systems. So we are talking about people who have petabytes of data, billions of files spread across hundreds of systems. So that's kind of been a pretty exciting thing. And it's a very unique set of challenges both for customers and for us to go so. Yeah, so it's interesting just to step back for a second. You know, object storage is something if you talk to anybody that's, you know, storage technologists, they're like, absolutely the way we need to architect things. But usually we tend to get away from talking about object storage itself. And it's really what do I do with it? What are those applications? What are those use cases? So there's still object underneath, if I understand it right. It's just you're getting closer, moving up the stack a little bit and getting closer to what your customers were asking for. Absolutely. The underlying infrastructure is still a collection of cloud services, not just object and S3, but a bunch of other services which are very API compatible with the cloud. But really that doesn't matter because those are just tools. What matters is what are you doing with that? And what we are doing to begin with is really backup archive and discovery of massive files inside the enterprise. All right, so, you know, there's some that we look, you know, backup we've been doing for a long time, but, you know, backups have been broken. We were at the VM World Show. There was a lot of buzz around some of the new companies. Sometimes they call them secondary storage, you know, rubric, Cohesity, you know, Veeam who everybody knows from the virtualization wave. Why don't you tell us, you know, are you part of kind of a similar wave? How do you kind of compare and contrast to some of those other players? You know, great question. It's similar, but quite different. If you look at rubric or Veeam, for example, Veeam really came about by doing tight integration with Veeamware and doing a Veeamware specific backup, which was the right technology, the right time for VMs and virtualization. Similarly, rubric and for that matter, Cohesity are really reimagining data protection primarily for structured workflows, databases, physical servers, VMs, tightly integrating it and reimagining how that feels from an experienced point of view. We are really looking explicitly at unstructured data. This is data which lives on network attached storage devices from a NetApp or a Dell EMC or a whole bunch of others. And the content is really digital assets. It's data that could be media data. It could be, you know, microscopy imaging. It could be design data for a variety of workflows. And this stuff continues to grow. It's monotonically increasing in every place, whether it's on-premises or on the cloud or at the edge. And protecting and managing this data is really a challenge and getting worse for customers. Yeah, the word that keeps coming up a lot, it's data. And one of the things I know we've been excited about, storage used to be about storing it. Now when we talk about data, how do I, you know, leverage it? How do I get value out of it? How do I discover different pieces of it? How have you been seeing things changes? Your background, you worked on some of the, you know, scale out NAS solutions in the past. So, you know, how do we see, you know, kind of unlocking the value of data? You're absolutely right. If you go back 10 years ago, the real problem was, how do I store all of this data? Today, there's plenty of solutions for where you store data, especially on the primary tier, right? The challenge is really getting data from where it lives to where it's needed, whether it's backing it up or archiving it to the cloud, being able to automatically discover things about it, simple things like, how is it growing? Who is using it? How big is it? How much of it is? What size of data? What about things that you can infer about it by looking at the type of data it is? This is what now becomes valuable because if you look at the data sets and sizes, even modest-sized businesses today will have petabytes of data, billions of files, and that's challenging for any system to go sort of understand unless you build it as a part of the platform. Okay. How about organizationally? You know, one of the other shifts we've seen is, you know, it used to be the storage administrator. How do I write? How do I grow? How do I manage it? You know, how do I have, you know, all of my protections and things set? A lot of the type of applications you're using are closer to the business. This is what runs the business. You know, the business user needs to be involved. How are you, you know, setting your solution up to, you know, do what the business user needs? Great. Yeah, that's a good question. Today, if you look at this data sets, this is not stuff that is an IT application. It's an end-user business-focused application whether it's research in a life sciences world or it's designed in a electronic design world, right? And in all of these cases, essentially the end-user cares because this data is critical to their daily working experience. And now IT is clearly involved. It's a clear sort of partner of the business unit in actually operationalizing this data and making it easier to go consume. But now it's really a joint thing. The final decision-maker is always the sort of end-user. In fact, we find ourselves in multiple places where we talk to IT and talk to the IT teams. They get excited, but very quickly they bring in the end-users to make certain whether the end-users are researchers or software developers or even hardware developers to make certain that they are comfortable with what we are talking about and they get really excited and that's the sort of starting point for our deployments. Yeah, we saw similar dynamic between the business and the IT when we talked about cloud. And when I talk cloud, I specifically mean public cloud and your customers, I have to imagine, they're all using public cloud in one way or another. Maybe explain that dynamic, how public cloud fits in with what you're doing in some of those IT and business people. Right. Look, the cloud is simply the most disruptive trend in the last 10 years, right? In fact, you got to go back to VMware and VMware's virtualization to see another trend of that magnitude. And all of our customers are embracing the cloud. They are wanting to go adopt cloud patterns, if you will. But the one area where they are massively challenged is around large data sets. Think about it. If you have petabytes of data that continues to grow, it's billions of files. It's spread across multiple geographies and dozens to hundreds of systems. It's a challenge to go leverage this in the cloud. So they are looking to us to be able to go chart the journey from all-on premise to a true hybrid world where they can use those cloud patterns much more effectively. Yeah, I'm curious. And maybe it doesn't fit exactly for what Igneous is doing today. But we've been talking about the data center kind of versus the public cloud in a lot of those environments. I talked to some companies that when I'm building those data lakes, I'm doing that in the public cloud too. Then the discussion that's come up a lot in the last year is edge. So IoT applications, we know we're going to have orders of magnitude more devices and there's going to be a lot of data. But the requirements for the data center versus the public cloud versus the edge are very different. How does Igneous look at that? How are you having those discussions? Customers, how do they get their arms around all the various places of data? Right, you're absolutely right. The requirements are different as in the public cloud is this massive hyperscale always available. The enterprise is a smaller version of that and the edge has a very different physical characteristics. But what we believe is important is the same patterns, the same APIs are available everywhere. And if you look at what the big public cloud providers are doing, Amazon with Snowball and Greengrass, they're trying to go move their APIs out and we completely embrace that trend. And that's one of the reasons we built our platform to be API compatible with the cloud with a variety of the cloud services because that means that the services we run can run in the enterprise data center or in the public cloud or at the edge, all on a platform which is appropriate for the three areas. Yeah, and to drill down specifically, you say API compatible, that's S3 really compatible. And do we have API creep? Every cloud seems to have, you know, not only one API but many APIs especially, you know, our friends at Amazon. You know, what are you seeing out there and what is the breadth of offering that you have today? So, it's S3 as a content storage layer. It's the obvious one, but the ones that we did not talk about the last time were things like index store, right? So, this is the equivalent of Amazon's DynamoDB or Azure stable store, the ability to go store a massive amount of index. But it's not just that, it's also the ability to go run compute close to this data which boils down to Kubernetes and containers. So, all these three are part of our underlying platform. We don't talk about that to customers except after they become customers, we really focus on the application which is backup, archive and discovery of all of their file data. Yeah, Kieran, take me inside to the customers you are talking to. A lot of times we're like, I hear this term secondary storage out there. And, you know, I worked on like converge and hyperconverged stuff. You know, those terms are something that customers hear about after a while but they don't solve the problems. What, you know, can you help translate for us what's going on in your customers and why is secondary storage important to them? What's different than traditional backup and how do you fit in? Right, so if you look at all of these guys, the data, the fundamental truth is data sets are growing and they're growing monotonically, right? Every year it's more. We've talked to folks where in the two years that we've spent as we were growing up as a company, they've sort of essentially had a 40% growth in their unstructured data sets, right? So then the question is a couple of things. One, they clearly realize that not all of that stuff needs to live or should live on high performance relatively expensive primary tiers, right? That's the first set of piece. But the question is, how do you find out what is active, what is not active and how do you move it to the appropriate place? So this is the sort of trend line and this is the patterns that they are living with. What we do is go in very simply, start off by saying, let's go find all of your filers. You know some of them, some of them you may not even know about and let's go automatically back up all the data and give you intelligence about that. What's the sort of simple intelligence? The intelligence could be how frequently are these data sets changing? How frequently are parts of this data being accessed or modified by your applications? So that's sort of first part of this and what this drives to is not only does it reduce the cost of backup which is really an insurance policy, it makes possible a bunch of intelligence about the data itself which is the beginnings of sort of appropriately staging data on the right infrastructure. All right, Kiran, you've had a number of customers since the early days. Talk to us a little bit about the journey you've been going on with them. You know how many of them have been pulling you towards the direction that you're now going? What's their response been? So I guess we call it kind of storage as a service where you're today. You know people love the whole concept of our offering as a service. Initially when we talk to customers they're kind of a little skeptical of our ability to go through this, but they very quickly fall in love with that. It's pretty amazing. What's not to like about infrastructure that is inside your data center but that you do not have to manage at all? And when I say do not manage, people don't even look at things like drives or CPUs or networks. That's not the world they live in. They live in the world of what's logically important to them which is on my backups running is my data being archived? How quickly is my data growing? Who is accessing this data and so on? And it goes to the next level which is they don't have to go manage things like software updates. Just like you don't know what version of Gmail you're running or you do not know what version of S3 is being used in the cloud. Our customers don't know what version of this is API level compatibility and there's a guarantee the services are not interrupted and they absolutely love that aspect once they get used to it. We tell our customers you don't call us. We call you if there is an issue and we're living up to that and they're pretty jazzed about that. I love that the version control thing is something that we said is something that cloud experiences actually want. At Wikibon when we wrote true private cloud it was like exactly that. You don't know or care what version of Azure you're running. You assume that they're going to test that out and do that. Can you give us any kind of concrete examples? Customers love if you can share any names but a lot of your customers are quite big. What are the concrete results? What are they seeing and any good stories you can share? So I'll give you an example of one of our largest customers. Can't mention the name but there's a large tech company in California. There's a lot of large tech companies in California, isn't it? I didn't tell you one. It's a story from Palo Alto, there's a bunch, yeah. Well, let's go further south in California. And these folks had an enormous amount of data and we started off by telling them, hey, give us your most complex systems, the ones that you are not able to go back up today. And we started with their file systems which were literally had this thing called file density which is an enormous number of files in a relatively small amount of storage. So you're talking about a billion plus files in terabytes of data and this is things that they had never been able to back up. And we go off and we are able to go back it up and consistently protect. So that's an example of a use case where we can go to a customer and allow them to accomplish what they cannot do today. Just from a basic backup point of view and take to the next level. In fact, they did this great demo for their internal teams where they showed how easy it is to search through this data and essentially accomplish in seconds what's typically in their current world takes hours to do. Okay, that's great. Yeah, sounds like you have some really good, interesting, large companies there. Is that, what's the typical profile you see? Is it really companies that have specific challenges because they've got the massive scale? How far down does this scale? So, that's a common question which comes along and the way I like to answer that is we are applicable to people with lots of data. It turns out they could be much smaller companies with lots of data. So we've got customers who are in the hundreds of people only worldwide, maybe two or three locations but they are really looking at a multi-petabyte sized data problem similar data density problem. In fact, another one that we are working with has got 300 million files and a terabyte of data. How do you back it up? How do you go discover information about that? That's what we solve and for these smaller companies which still have the problem, they are actually starting to find out about us and come to us which is really gratifying. Okay, well you seem pretty excited about the space. What's exciting you the most about where we are today with the technology? So, the real issue is people talk about data and they immediately go to databases, they talk about virtualization and physical servers but that's not where the data lives. The data hasn't lived there for over a decade and more and more of the data lives outside in files in objects and there is this sort of ability to go, understand that better, manage that better, protect that better and last but not the least sort of provide intelligence to users because this data is something that they care about. People are not keeping this because somebody else told them so it is their life plot. It is their sort of livelihood if you will from a company point of view and helping customers be able to go take that to the next level bring the sort of cloud patterns to these use cases that's pretty exciting. Yeah, absolutely. I want to give you the final word, right? I hear this and I think about the whole wave of big data what we're starting to talk about continuously with AI and ML really it is about unlocking data so huge opportunities going forward. Any of the other trends outside of what we've discussed already that you want to give us for a final word? You know the last thing that I'd say is it is about data, it is about complete automation all across the stack whether it is storing, managing or deriving intelligence and the reason you want to go automate that stuff using intelligence in the software systems itself is simply because it's too large there's no other way to go do it and last but not the least all of this stuff has to be offered as a service because the cloud has gotten people really hooked on this sort of comparatively easy world of not having to go manage infrastructure and I think those are the three things that we should be hold by. All right, Kieran Baggespore, really appreciate the update on igneous systems, absolutely customers dealing with massive amounts of data. How do I unlock the value of that without having to be down in the guts which has been really the history of storage? I'm Stu Miniman, thanks so much for watching theCUBE.