Episode Transcript
[00:00:00] Speaker A: Storage. Whilst it goes back over 20, 30, 40 years, it continues to show itself as being that kind of critical part of the infrastructure that has huge value once you know how to integrate it and unlock it up into these new environments that people are building and working on.
[00:00:21] Speaker B: This is kvcads as a primary target for ransomware campaigns, security and testing, and.
[00:00:28] Speaker A: Performance, risk and compliance. And we can actually automate those, take.
[00:00:32] Speaker B: That data and use it.
Joining me back on the show is Matt Watts, who is a global data expert. And today we're discussing the journey from storage to intelligent data infrastructure. Matt, welcome back.
[00:00:46] Speaker A: Marissa, it's great to be talking to you again. Thanks for having me.
[00:00:49] Speaker B: You know, it was interesting before we jumped on to actually do the recording we're talking about. I'm like, I feel like I just interviewed you and I just saw you, which was six or so months ago now. But when I did see you, you wrote a book, which I read on my very long trip back to Australia from Vegas. So maybe let's sort of start there, give people a bit of an overview of what your book's about.
[00:01:09] Speaker A: Yeah, sure thing, Carissa. So let me start off by saying that it was never intended to be a book. I just had all this kind of stuff in my head a year or two back and just decided I would start writing it down. And I didn't have a purpose for it. It might have become a blog post or it might have become a series of LinkedIn posts. And after spending time, it's like weeks, I suddenly realized I'd written pages and pages and pages and pages. And as the more I looked at it, the more I started to realize there was a pattern emerging from it. And at the time, it was the third wave, because we'd sort of see these three big waves in technology, and that was kind of how the whole thing started. And then as we started to see the emergence of AI, I suddenly wrote some more about that as well. And that kind of became the fourth wave. And, you know, to take a long, very boring story and make it short, at that point, it just felt like there was enough there for us to put it out into the wild and just see if other people connected with it, resonated with it. And we've had some really good feedback. It's been. I've been told it's.
Some people have said to me, it's like actually listening to me. It's like having me sort of talking to them, which I think is quite a nice compliment. So that's how it came about.
[00:02:17] Speaker B: Yeah. You've definitely humanized it, especially with the little images and, you know, some of the language that you use in the vernacular. So. And it was short, so it wasn't like you've created a book that's 400 pages. I think it was like 80 or so pages. So I think that it was short and concise, given the title.
Let's walk through that. You've just mentioned the waves, but I'll sort of give a bit of a. An overarching view of the waves, and I'd love you to sort of get into a little bit more. So the first wave that you outline in the book is modular technologies, followed by the second wave, which is virtualization.
Then the third wave is cloud computing, and now in the fourth wave is the Intelligent Data Infrastructure. So you sort of talk about your timeline as well as your pedigree in this space to say that you've been through all of the waves.
So maybe do you want to step through each wave and then just give a little bit of context on, I would say, what that looked like in that moment and then how it sort of maybe started the second wave and the third and so on.
[00:03:16] Speaker A: So it's very much kind of based on my career and my sort of history within it. But I think the whole concept of visualizing things as waves is quite an interesting construct that you can take into lots of different industries. So I think people I've been talking to over the last few days have sort of said, oh yeah, I can, I can see waves in my industry.
So this goes back to sort of the beginning of NetApp, which, the first wave we just talked about, which was modular technologies. There are going to be so many people on this call that will not remember the technology from the late 90s, the early part of the 2000s, and it was a world of monolithic technologies. We had IBM had the Shark and HDS had the Lightning and EMC had some metrics, phenomenal pieces of technology, but incredibly rigid in terms of how they, they performed, how they behaved. And NetApp was born into that industry and it wasn't a question of coming in with a slightly better version of what is already there. Our idea, our concept, our sort of raison d' etre at that time was that we think there's an opportunity for a wave. And the wave that we saw was this switch from those monolithic type systems of the past towards these more modular type technologies. And we believe that was going to be the way people would do things for the future.
[00:04:33] Speaker B: Would you say if you were to zoom out across all of the waves. Just for a moment. You mentioned the word, you know, being quite rigid. Would you say that now with how things are, technology is not as near as rigid well, as it used to be. But then also there's a lot more flexibility that we're starting to see come into play.
[00:04:49] Speaker A: Yeah, entirely. I think it's exactly that, is that the way that people benefit from technologies is when we make them more accessible, we make them more flexible. We use technology to give the business much more sort of opportunities to do things faster, to do things that they couldn't do before.
And that was really why that first wave was so successful. You know, suddenly we had the Exchange team building the Exchange infrastructure, the Oracle team, the Oracle infrastructure, the SAP team. They were suddenly able to bring new applications into the business using these modular technology stacks. And that created huge innovation for the different teams within the business. Suddenly there were new applications coming in that they could use that was creating new levels of innovation.
And I think that's symptomatic or it's a cause of every one of the waves. I think it enables those kind of technologies, those kind of capabilities within the business, and that's why these waves form and build such sort of momentum and such volume.
[00:05:48] Speaker B: Okay, so then let's just move on now to the second wave. So virtualization.
[00:05:53] Speaker A: Yeah, I mean, the virtualization wave, you mean, probably started in around sort of 2008, something that sort of timeframe. And there'll be people again listening to this that don't remember the world before virtualization, but it existed and it was pretty horrible. The first wave, whilst it solved a lot of problems, it created a lot of opportunities. As that innovation started to slow down, it also gave us an awful lot of inherited complexity. And that was kind of why virtualization started. It was a way of solving some of the problems that we'd actually created in the first wave. A lot of underutilization, there was complexity skills, and virtualization changed everything.
It's hard to help people understand just how fundamentally the landscape of it changed. And it wasn't always just VMware. We had Citrix and VMware and Hyper V and KVM and lots of different technologies. And over time, I think most people sort of consolidated that down to VMware. But it goes back to what you just said, flexibility.
Suddenly we were giving the ability for people to spin up new applications, to create automation, to do workflows. So virtualization, it sort of started as a way of simplifying and creating a better level of Utilization. But what really got the wave going was that automation. It was that ability for people to do things at a speed they'd never been able to do before.
[00:07:15] Speaker B: Okay, there's a couple interesting things in there that you mentioned. So you said get the wave going. So would you say now, like, the wave's really, really going? Because like, every day I'm in media news coming out, there's a whole new technology that's emerged from somewhere we never even heard of, and now they've overtaken the, the original player.
So have you also seen as well that, you know, you said before that, you know, the rigidity in the first wave. Would you say that? We've been carrying a lot of this, you know, legacy technical debt along the way?
[00:07:44] Speaker A: So I think we, I mean, there's always, there's always kind of legacy technical debt. You know, our job is to try and make sure that we're not spending too much of our time, energy focus, being consumed by that. That's stopping us from kind of moving forward and embracing technology and innovation. You know, I think the waves don't go away, you know, as, as a new wave starts to form. It doesn't mean that the old one simply stops. It's just that what happens is that our focus tends to shift because we tend to see that the biggest opportunities are the ones that are presented us by this new wave that's forming. So, you know, there's always going to be technical debt. I've said to people in the past that, you know, quite often for a lot of companies I talk to, they're not really data centers, they're museums of past IT decisions. Because once we start embracing technology, often it's very difficult to phase it out or to carry it all forward with us. And sometimes, you know, we struggle with that sort of legacy technical debt and we just have to be aware of it, conscious of it, manage it, and in some cases, let it go where.
[00:08:48] Speaker B: We can, would you say? I mean, especially because I'm, I'm an ex, I worked in a bank, for example. Because I've obviously got, you know, they're very old and they've got a lot of, you know, files and systems they have to keep running. Would you think it's harder for certain industries to be able to sort of let things go?
[00:09:04] Speaker A: It's not, it's not even just industries, Carissa, which is really interesting. It's in. It's human nature. You know, we tend to be hoarders. We don't, as a, as a species, we don't like to get rid of things. We don't like to let things go. I've literally just come from meeting with a client at lunch and we were talking about just looking at data. And the reason we got talking about it was because recent statistics show that about 68% of the data we create is never ever used again. After it's created, it's just single use data. So you'd think, well, that's a great opportunity, let's get rid of some of that stuff. But we struggle to, you know, we struggle to work out who owns it, who's responsible for it. And heaven forbid if I did delete something because I don't believe it has value. What if somebody in the year, two years, five years time suddenly says, whatever happened to that piece of data? Because that might be useful. So we do struggle, I think, generally to let things go because we have that fear of what if we might need that in the future, or what if something means that was more important than we realized in the future. So I think it's a really interesting kind of challenge. It's not just a technology challenge sometimes it's also that fear of can we release this technology, can we let this thing go? Or is that going to cause us a bigger problem?
[00:10:19] Speaker B: Yeah, no, that is interesting. And I guess, would you say from your experience that to your point that we don't know who owns it and it's probably someone who did own it, left there 10 years ago and it's just sitting there somewhere being stored, would you also then think that people, especially larger companies, do they not want to make decisions to be like, okay, like you said, there's 68% of these things that have been used once. We probably don't need all of it.
How would people start to work through, like prioritizing what to get rid of and to know what to keep and to know that, does anyone even own this at all? Do we even need this?
[00:10:51] Speaker A: It's a data management challenge. You're absolutely right. And now that we're starting to see this kind of quantity, companies are recognizing whether it's from an energy perspective, a cost perspective, a sustainability perspective, that there's a really good opportunity through better, more effective data management that's a great candidate for us to be able to address energy sustainability and costs. But it's really difficult. I mean, one of the tools that we provide to companies is something we call BlueXP classification, which is effectively a data analytics tool. It allows them to scan all of the data they have, understand what is it who created it? When was it created? When was it last accessed? Does it contain things that are sensitive?
So we can give people a lot more of that information, which I think is gonna underpin any strategy that they could put in place to work out how to better manage.
But I think it's going to need partners, it's going to need probably some outside help to really look at that data and help people say, you know, how do we effectively manage this over a longer period of time? You know, is there a better media we could store it on? Could we move it to the cloud? Could we, heaven forbid, could we delete it? So I think there's an increasing awareness of just how much of this legacy data we're carrying and dragging forward with us. And there's a realization that there's an opportunity there. And I think just combined with that, it's going to get worse because now we're seeing the next waves forming of cloud and AI and we're creating vast quantities more data. So that's only going to pile up into that kind of, that debt of information that we carry. We keep lugging forward with this.
[00:12:29] Speaker B: Do you think as well, going back to data management, you said that obviously that's a massive challenge, but do you think it just is one of those things that sort of just gets pushed down the laundry list because again, there's other things that there's always going to be something else that you want to be able to do. And again, it's a big task. It can feel overwhelming, especially, you know, the bang. Like some of these places, 150 years old, right? Like they've got these legacy systems. You don't even know where it is, who's got it. Do you think that people just couldn't be bothered, it's too hard basket? So what we'll do is we'll say we'll do it end of the year and then we'll just push it into next year's agenda. How does that conversation go?
[00:13:01] Speaker A: So I'm sort of sniggering because I've lived in and around this world for almost 20 years.
I think that is the reality of it. It's that we rely on the fact that technology progress will allow us to push the problem out. And we've done that every year. Every year, Every year, hard drives or solid state drives will get denser, cheaper, we'll get better compression, deduplication, compaction, we'll get better efficiency technologies. And I think that's what people do, is people look at it and say, well, I'M okay, maybe even spending a little bit more if that means that I can just push out this problem, because technology will solve most of it for us. But I think we're getting to this sort of interesting point now where our ability to create data is so far beyond the progression of technology that it's now going to become something that people are going to have to take more seriously. They're going to have to start really looking at how do we deal with this.
[00:14:01] Speaker B: Enterprise tech leaders know that compliance isn't just about ticking boxes. It's about risk, reputation and revenue. That's why companies trust Vanta to streamline their security and compliance workflows at scale with deep integrations and automated evidence collection. Vanta takes the manual audit grunt work out of the frameworks like ISO 27001, SoC2 and GDPR. Visit vanta.com kbcast v a n t a dot com kbcast to learn more.
Okay, so then let's move forward now into the third wave cloud computing.
[00:14:40] Speaker A: Yes. Yeah. So without a doubt, I think, you know, cloud for me is definitely the third wave. You know, the latest kind of estimates are we spending about 600 billion on cloud. That feels like a bit of a wave to me. And it has had a profound effect on the way that we operate, run, manage and deal with it. So. And third, you know, netapps had a role through all of these. You know, even in the second wave of virtualization, a lot of the kind of capabilities that we have, we can snapshots and cloning, we could expose those up into these virtual worlds and bring these virtual worlds to life or to create really new and often exciting opportunities with cloud. As people have started to look at how do we create hybrid cloud, you know, how do we tier, how do we back up, how do we, you know, replicate into the cloud? That's kind of where it started. Then it was how do we start to migrate workloads into the cloud? And now it's how do we bring in tiny new workloads online inside the cloud? So you can definitely look across most organizations and they are, you know, at some stage of that kind of cloud journey and working out what role it's going to play within their organization. But when a market gets to 600 billion, that is without a shadow, without a wave.
[00:15:52] Speaker B: Yeah, and I think somewhere in the book you mention, and I mean I've spoken about this too on the show is when the Internet came out, everyone's like, oh, that's not going to be a thing. And I think you Say in your book as well, for memory, like, people sort of shrugged off cloud as well. Sort of putting into the same bucket, like it won't take off. Which was, to your point, look how big the market is now for cloud.
[00:16:10] Speaker A: Yeah. I mean, again, it's different. Different companies do it for different reasons. Some there was. I think there was a fear of cloud to begin with. You know, a lot of people felt that it was going to take their jobs away from them. And to be fair, I think that's a natural response to any wave when something that significant happens. Our inbuilt sort of fight or flight response is, is this going to have a negative impact on me or is this going to change something for me? And therefore I naturally feel like I want to push back against it.
[00:16:37] Speaker B: Yeah. So, Matt, I now want to move into the fourth wave. Intelligent data infrastructure.
[00:16:44] Speaker A: Yeah. So the fourth wave actually is to AI. And that's this idea of we want to help companies build intelligent data infrastructures because we feel that's necessary not just in support of the waves that have gone before, but definitely in support of where things are going for the future. I mean, AI for me was a really interesting one because whilst it's not a new phenomenon, I mean, oh, my God, if you go back far enough, you know, Alan Turing was talking about, you know, if you could, you know, the famous Turing test, if you could. If you could talk to a machine and the conversation was indistinguishable from that of a human, then you could almost class that as a thinking machine. You know, that back in the 30s and 40s. But what kind of happened with an it? Well, when I talked about the third wave, the very first personal question people would ask me is, you know, well, what's the fourth wave? Or they would say to me, do you think the fourth wave is AI? And it was probably for the last two years, I kind of said possibly because it wasn't that we didn't have the technology.
You know, if you think about AI, you think about what we've got with Nvidia, you think about what we've got from a whole bunch of different vendors. The technology has been there for us to really do some really interesting things with AI. The challenge has been the skills that up until late last year, there were a very limited number of data scientists, of people who could really bring AI to life inside their organizations. And if you can't do that, if you can't democratize something, it can never become a wave. And then, of course, in November last year, we had the ChatGPT phenomenon and we went from a. I think it went from like a hundred thousand users to 200 million users in a couple of weeks. And suddenly AI had been democratized. Suddenly the programming language for AI was the written or the spoken word, and that is what started the arms race. And everything now is around AI. If you, you know, even at a government level, every country, every government, every company is now looking at how do we take this AI thing and use it to create these, these huge advantages or, you know, to really live out what the promises of AI is going to be. So it's definitely a wave. You know, you only have to look at the US I think it's 500 billion into Stargate.
The amount that is being pushed into this technology is just absolutely staggering. So I think it'll be a, a very interesting year or two years now as we start to, to see the, the manifestations of this and what it gives us the ability to do.
[00:19:08] Speaker B: Yeah, that's interesting. So then, so then going into this a little bit more, would you say that the whole AI wave is gonna be the fifth wave now? Would you say? And I know, like, I sort of jumped ahead, but just as you were talking then, and you're right, AI has been around for, for longer than people would realize. However, I would say.
And would you agree when ChatGPT really came out, what, at the end of 2022, would you say that was sort of the start of like the hardcore huge tidal wave coming in now in terms of how ubiquitous it then became?
[00:19:40] Speaker A: Yes. I mean, when, you know, when, when the ChatGPT launch happened, you know, suddenly, suddenly people were, could see the possibility then. That's the other thing as well. It wasn't just that it gave us an easier way to access it. It became, it went viral. Right. I mean, if you think about it, everybody was suddenly using ChatGPT and finding. Use cases and you know, when, when something goes from that, that sort of obscurity into that sort of viral possibility, it's created this, this really interesting world where suddenly everybody, inside departments, inside businesses, inside countries, they're all looking for how do we use this technology now to create opportunities, to create new possibilities, to create new innovation. But you said something interesting is the fifth wave. Knockhead. There is a fifth wave. I can kind of see it, and I got a good feeling as to what I think the fifth wave is likely to be. I think the fifth wave will enhance what we've had with AI. I think it will create even more momentum with AI. But I actually think there's a fifth wave on the horizon and it is standalone. It will help with things that have gone before. But I can see something coming that, that is, it's very exciting for me.
[00:20:55] Speaker B: So in the book you sort of say, I'm going to read it out to you. Say I want to start out by saying that these waves do overlap. Don't think that one finishes and then the next one starts. So how do you sort of know? Like conceptually you sort of know another one sort of starting up and it's hitting a little bit harder. Like we've just given the example for, you know, Open AI and Chachi. But what about some of the previous waves? They've obviously cycled down a lot more and I'll get to that in a moment. Other parts of your book that you cover. But do you think it gets to a point where it's like that waves completely dead or what, what does that look like?
[00:21:26] Speaker A: You know, I think if I had, if, if I could go back beyond the sort of the first wave. I mean, because the first wave is really the waves as, as I've seen them within my career that were probably waves before. They're just not something that that was my career was really focused on. But I think just looking back at history, nothing ever really goes away. I can remember mainframes, you know. Are mainframes dead? No, you know, mainframes still sit there. IBM has a very, very healthy Z series business selling mainframes into some of the large banks and finance institutions where they sit at the core of what they do. Tape. Not that I'm saying tape was a wave, but you know, even a technology such as tape, I mean, tape's been dying for the 30 years that I've been in the industry. And it continues to find a role for itself. It continues to find a use case for itself. So yeah, these things have a very, very long tail. And definitely the four waves that I can see within my career, people are still buying modular technologies to support whether it's virtualized applications or enterprise applications for on premises workloads. People are still investing into virtualization to virtualize certain parts of their environment, or they're investing in virtualization and the cloud providers. Cloud computing continues to grow at a pretty staggering level.
So yeah, I think what I can see if I look back at history is all of these waves and the technologies that sit behind them have a very, very long tail. How do you spot a new one? How do you spot when one is forming?
I wish I had the Formula. I think you're looking for a lot of different signs, a lot of different patterns and even world events. You know, certain things sort of come together and, and, and if you've got enough of the tea leaves and you, you've got the, you know, you've got enough intelligence, you've got enough, and no genius luck, it's a sprinkling of all of those things. And so it's, it's not easy. And the AI one is a really, really good example, you know, that we've had AI and people have been experimenting with that for two, three, four, five years. But it took a trigger. And the trigger event was ChatGPT.
That was the moment that suddenly we saw the democratization and it kind of came out of nowhere. And I think that's, that's what, why this all fascinates me is that you can kind of, everything can line up and you can think this has a possibility, but then it needs that one more trigger to happen and it can suddenly happen out of the blue without you realizing it. And that's that, I think, is why all of this fascinates me so much, is because I'm always looking at what might the next one be. And I've usually got three, four, five, six different things in my head of what I think the next wave could be. And against each one I sort of have all these characteristics of this has happened and this has happened and these events have happened. And so I'm always kind of assessing, you know, is this good enough? But it's sort of waiting for one final thing to happen in order for it to trigger. And I'm not, it's not an exact science, but I think I've got a good feeling for what the next one could be. So, you know, I'm sure we'll come to that.
[00:24:34] Speaker B: I do want to get into that. I want to go back a second and get back on tape. So I interviewed someone the other day that said Taped is making a resurgence apparently because of ransomware. That's like, oh, you know, if something's backed up somewhere, et cetera. But there's a ransom and people don't want to pay it, for example, or they get it back, but they don't have everything back. Apparently people are resorting back to tapes. Have you heard this?
[00:24:56] Speaker A: I must admit I don't tend to get involved in too many tape based conversations just because of the kind of audiences that I speak to. But it wouldn't surprise me. It is one of those technologies that, you know, is just continues to, to find some new use case, whether it was storing, you know, very large data sets, because, you know, actually streaming very large data sets onto tape and reading them back is actually quite quick.
So it would not surprise me at all. There's been a resurgence because it's now found a sort of another use case because of these, these sort of world events, world scenarios, world or ransomware attacks. So it wouldn't surprise me.
[00:25:34] Speaker B: Well, I mean, just to go into this a little bit more like, I was shocked. But then I asked, I'm like, the theory checks out, as I said, like around, you know, the backup and if there's a ransom, people don't want to pay it or if they even get full, you know, all of their data back, for example. But I said, yeah, but then it unlocks another problem that, and I've been in a company before where the tapes went missing, or then they gave an example saying that, you know, people would collect the tapes from the company to, you know, presumably back them up in a, in a safe place. So they'd come and collect it, but then they'd stop over and someone would copy the tape. So I think if you're opening up a can of worms either way, but I was shocked by that. So I hadn't, I haven't really heard tape as a backup for like 15 years in sort of that capacity. So when you mentioned that, I thought, actually I've interviewed someone recently that spoke about tapes.
[00:26:18] Speaker A: You know, if I was to look at it myself, I would have thought there's, there's other ways that you would, that are probably more effective in terms of recovery. I mean, if you can lock down your, if you can lock down volumes of data, whether that's on premises in the cloud by using kind of worm technologies, you know, write once, read many, then effectively that becomes a very, very secure recovery point as well. So I could see, I can see why somebody would say that there's a use case for tape.
I can also see that there are very good arguments to say actually backing up into the cloud, backing up onto some locked type media also is a very effective use case for protecting against ransomware and may actually offer benefits in terms of the speed of recovery and things like that as well. So I can understand it. But I could also see arguments that there are still other ways that might be just as effective, in fact, you know, if, if not, maybe more effective.
[00:27:09] Speaker B: Okay, so now I want to go to another part of the book where there's a graph and I'm going to sort of Talk through it, because obviously it's, it's an audio interview.
In the graph, you've got three bubbles. The bubbles are infrastructure, applications and data. And you've got a bit of a timeline. So starting from around 2000 up till about 2020, so the bubbles in those area, as the time sort of goes on and elapses, the bubbles change in size. They're still there, but the focus sort of changes. So how do you sort of weight those bubbles of the infrastructure, applications and data? How would you sort of weight them? Now we're in 2025, so it's kind of interesting.
[00:27:47] Speaker A: I can see the picture in my head because I've spoken about it so many times, you know, and it does sort of line up with the waves. You know, if I think back to the first wave, it was a. It was a very much an infrastructure wave. And therefore the conversations I was having with people would be very big, kind of let's talk, or very much focused around infrastructure. Then there would be a bit of a conversation about the applications that run the infrastructure. And then data was almost the third most important thing, or it's the last thing you sort of got to. And then as we went into virtualization, people tended to be less interested in the infrastructure, more focused on applications, because that was where a lot of VMware's value was, was the sort of application automation.
And data suddenly played a much more significant role, and for two reasons. One, people were realizing there was more value within the data. And two, virtual machines are data, virtual machines are files. So suddenly there was a recognition that data was more important than we'd realized because it was defining so much more of the environment. And then as we got to cloud, in the cloud, people aren't thinking about what the infrastructure is underneath it, per se. The number and the quality and the opportunity to test so many different applications became key. And I think data really blossomed. Suddenly people realized, if we've got our data in the cloud and if we have access to this huge number of applications, what could we do with that data? How could we create value from it? How could we learn from it?
So it really shifted from being big infrastructure focused and little bit of data to being this big data focus with much, much less infrastructure. And if I was to draw that picture now as we start to look at this kind of fourth wave, the fourth wave for me is we are in an era of data and intelligence and every conversation now starts with data. What can I do with it, what do I need, how can I refine it? To get value from it. How can I apply large language models to it? How can I. So I think now those bubbles, if you were to show them now, would be a massive data bubble, would be a significant application bubble. People are very interested in am I going to use Vertex or SageMaker or, you know, so they're very interested in the AI frameworks that they could apply against that data. And then there would still be an infrastructure discussion. But I think it's, it's, you know, it's still there, but I think it's less of the discussion. So long answer to the question. But that pattern continues with the data bubble getting bigger and bigger and bigger.
[00:30:16] Speaker B: Okay, so you would have remembered, I think around like 2012, 13, 14, obviously there was a big boom on let's collect every piece of data we can, because then we can create profiles on people. Because, for example, if I'm a bank and I can sell you a mortgage because I know everything about you, where you live, what your interests are, et cetera. So that's when, you know, Cloud, Dare and Friends were really sort of coming into play. So then now it's at a stage where it's like, while we don't want to hold that data from a security perspective. And then to your point, we're holding 68% of this data, which we've used once. Now we're trying to figure out a way on how to manage it correctly.
So where do you think it sits now on finding the equilibrium between. Of course, collecting the data. Data is king, we all know that. But then managing it from a, like you said, if you have too much that's sitting there, it's at risk from a security point of view. But then also it's quite costly, it's not sustainable, et cetera.
[00:31:11] Speaker A: I mean, it's interesting, Chris, you talk about that, because I think maybe it was just after that timeframe, but it was, you remember, the whole kind of data is the new oil. I can't remember exactly what year that was, but that was a sort of a talk track from probably the mid, sort of 2015, 2016, something like that. And it was, it was the time where the more of it we have, the more value that we can get from it. And I kind of get the sentiment that, you know, with oil, it's, if you can refine it, then obviously you get a.
The more refined, the more you have, the more refined it can be and therefore the more value it creates, just like oil, that there's a byproduct to it which is that, you know, we've got to store this, we've got to manage this, we've got to protect this, we've got to deal with sustainability, we've got to, you know, and actually what we want to do is be able to use less of it to achieve a better outcome. So I think with data, people are now starting to realize it's not necessarily the quantity of data that you have, it's the quality of data that you have. You know, if you're pointing a large language model at a massive quantity of data, but that data isn't quality data, then what do you think you're going to get from the large language model that you've built? If you put garbage in, you're going to get garbage out. So I think with LLMs, with AI, there is a realization that we do want a lot of data, but what we do want is a lot of quality data. Because if the better the quality of the data, the more value we can get from it as we start to refine it, using large language models, using things like that. And of course the byproducts of all of that are if we've got a quantity of quality data rather than just a large amount of it, then we are being more aware of how much we're spending in managing it, how much emissions are being created, because there's a lot of people who still very much care about sustainability. And of course you can then look at better data management as well. You know, if you're bringing together a smaller quantity but of larger quality Data. Are you 68% unused data? I would argue probably not. You've almost done a sort of a data management upfront rather than sort of trying to retrofit data management off the back end because you've now got so much of this single use data. So yeah, I think it, it's AI is definitely causing people to think differently about data and to start thinking that it is much more about not just quantity but also quality as well.
[00:33:31] Speaker B: And just before we move on a little bit more, I'm just curious now, as we've just discussed that people in an era where they were like vacuuming up all the data they could, like what we've just spoken about now, it's like, oh, GDPR especially where you, where, where you live, that part of the world that's in play with people trying to get rid of the data and shed the data. So now would you say, generally speaking, companies or enterprises more specifically are in this stage where they're trying to do the data Management, what data do we actually need from a, you know, managing gdpr, but then also from a security perspective and equally like you said, if they're trying to, you know, implement AI and they're running like so much, they've got so much data that they're running against an LLM, like it's not going to give them the most effective outcome. Right. Are people at this stage where they're just trying to look strategically at what is it that they really need to manage all the things I just listed before?
[00:34:24] Speaker A: Yes, I think so. I think people are now more focused on, okay, well what's the actual outcome that we're trying to get from this? And therefore let's, let's use the outcome to give us more intelligence about what it is, what's the sort of data that we want to try and get hold of. But it's really interesting when you talk about sort of regulations and I actually have met companies who haven't started on AI and when I ask them, you know, what's holding you back? What is it that's stopping you from doing AI? And they said, we have a fear of the regulations that don't currently exist. They said, what if we built a large language model? What if we built and it became a core part of our business?
And then there was subsequently some AI based regulation that said, actually you can't do one of the things that you're doing or you're not allowed to use that data or you can't. So a lot of people look at regulations and think regulations are bad, regulations are bad. But some, but that was just fascinating for me because that was a company who was saying, actually regulations are good because once the regulations are in place, we know what we can do with AI, what we can do that's going to be safe and isn't going to, you know, isn't at some point in the future going to have a negative effect? You know, where we are in Europe, we like a good regulation so we can certainly come up with them. But that's not always a bad thing. Regulations from a GDPR perspective, from a, you know, a hipaa and from all these different things, the way I look at it is they tell you what you have to keep, therefore everything outside of the regulations is an opportunity. You know, it could be that you say, well, if we don't have to keep it for any legal or regulatory reason, then it becomes a candidate for us to look at and say, do we actually need to keep it? Does it have inherent value? Not with relation to Regulations. So does it have inherent value that means we need to keep it for some other reason? So whilst a lot of people look at regulations, say regulations are bad, regulations are bad, I kind of have that opposite view that I think regulations kind of tell you, well, these are the things you have to do and everything outside of those things is kind of an opportunity, which maybe that's an odd way of looking at. Maybe that's a very emir kind of centric view of things.
[00:36:28] Speaker B: So now I'm going to flip over and again, going back to the book you talk about storage plays a role and it must be part of your cybersecurity environment with autonomous ransomware protection to talk us through that and what that looks like.
[00:36:43] Speaker A: So it's interesting that, and I sort of touched on it a little bit earlier on, but when I kind of think about storage and the kind of the role that storage has had to play. I came to work for NetApp because I thought storage was just where data went to live out its life. I didn't think there was anything there of intelligence. I thought the intelligence sat within the applications, within the operating systems. And it was like, you're telling me you can trigger an instantaneous backup of my data that consumes no additional storage until the prime data starts to change? Yeah. And by the way, we can enable you to restore it instantly as well. So that was my kind of moment. That was the monolithic to modular shift. It was the first time back in the 90s that I'd seen storage as something more than storage, as something that could have an important role to play within the technology stack. And then when it came into virtualization, that second wave snapshots kind of came back to life again. And suddenly, what if I could take a snapshot of your entire virtual environment to protect it from whatever? What if I could clone it? What if I could disaster recover it into the cloud? So suddenly storage, with some of its kind of existing primitives of snapshots and snap, restore and cloning, suddenly created huge value in these virtual environments. And then we got to the cloud and as people were putting workloads into the cloud, they realized actually most cloud storage have snapshots, didn't have cloning, or certainly not efficient cloning and storage came along again. We put our technology into the cloud and suddenly those primitives, snapshot, snap, restore, cloning, suddenly a new lease of life because they could deliver value into these entire application environments, people were bringing to life in the cloud and now we're in AI, and guess what? Similar things happening And a really good example is I had a conversation with a data scientist about two years ago, maybe a little bit less 18 months ago, and he said to me, you know, who are you and you know what a NetApp do? Said Matt Watts, et cetera, et cetera. I said, I work for NetApp. We help companies build intelligent data infrastructures. And he said, what does that mean? And I said, okay, well, let me put it this way. I said, what if I could build an intelligent data infrastructure for you that could give you a capability that you could trigger an instantaneous revision or an instantaneous version that you could hold, that would be for any data that you'd use to create or train a modeler going back over days, weeks, hours or months. I said, give you that history. And he said, that would change the way we do things. I said, what if that intelligent data infrastructure could clone such that you could take any quantity of data and create an instant zero overhead replica such that you could be training multiple models in parallel with multiple data scientists? He said, that would change the way my team operates. And I said, now what if you could do that the same way, whether it's on premises, in gcp, in Azure or in aws? We said that would change the way we do business.
That for me was just this wonderful example of how making the right infrastructure choices, in this case storage the infrastructure, whilst it's not the most exciting part of the conversation for most people, using the right infrastructure that has these kind of capabilities can have a performance profound effect into the environments that are built upon them. And that's kind of where I get excited. DevOps was a really good example of we'd had development teams or application development teams that had moved so far away from the infrastructure that they sort of did that glue, had all that connectivity had been lost and DevOps kind of came along as a movement saying, you know, what if we can make sure that the application and the infrastructure are working or are harmonized, look at how much better, how much more powerful the environment can be. And I think that carries on. I think DevOps will find a role within the AI movement as becoming that kind of glue between the AI applications and the infrastructure behind it. So, yeah, storage, whilst it goes back over 20, 30, 40 years, it continues to show itself as being that kind of critical part of the infrastructure that has huge value once you know how to integrate it and unlock it up into these new environments that people are building and working on.
[00:40:55] Speaker B: So, Matt, I sort of want to conclude our interview now with the fifth Wave. You mentioned before, you've got sort of five or six sort of points or a formula that you sort of look at. Is there anything you can share, considering tenure in the game, but also you've been through all of the waves.
[00:41:14] Speaker A: You know what, I guess I wouldn't look old. And this is just kind of, you know, this is some thinking about it and, but this is actually. And it's something that we're, we're actually actively working on as well with or a part of it. So we have, I don't think we've really talked about it up until now, but there's a relatively new phenomenon that people are starting to become very aware of. And actually you mentioned in the last question we were talking about, and I didn't really answer it was autonomous ransomware protection.
Security as a part of the infrastructure I think is absolutely critical.
And within the storage layer, having it dealing more efficiently with security and more autonomously around security, I think is absolutely critical. It's a very strong area for NetApp in particular. But the reason I'm saying that is because this new phenomenon that has started to sort of rear its head is this whole idea of steal now, decrypt later. And what I mean by that is that there is a growing realization that, that we are on the cusp of quantum. It's probably still five years away, but we're seeing huge innovation in terms of what we're able to do with quantum computers. We're solving for some of the scale problems. We're now starting to see error correction capabilities coming in. It is still, you know, let's say five years away. However, if you are, let's say you build military aircraft, then that aircraft takes you about 10 to 15 years to design. It goes into service for maybe 20 years and then it'll stay classified or top secret for a further 15 to 20. So that is somewhere between 50 and 60 years that your data needs to be secure, encrypted and top secret. If I am a state actor that can get access to high performance quantum computers as they become available, then I'll steal your encrypted data. And I will wait and I will wait for the five, six, seven, eight, nine years that it takes for us to have a quantum computer that is powerful enough to break the encryption we already know. TLS, IPsec, SSH, RSA, SHA, those are not strong encryption forms. They are essentially very weak encryption when it comes to us having quantum capabilities. It's the thing with blockchain. Blockchain's using elliptical encryption that is not a strong encryption. When we start to see a world where we have quantum computers that are powerful enough, with enough qubits for us to be able to solve this problem. So we now have to start thinking about how do we protect our data if we need it to be secure and encrypted and classified for more than five, six, seven years? We then need to think about what does it mean to be in the post quantum world. The second thing around Quantum is there's a capability called quantum entanglement, quantum transportation, which means we can use quantum as a new way of networking or of transporting data across networks. That's moving forward very quickly. And then third, quantum computing, when we have these computers that are powerful enough for us to solve some of these big problems, Quantum will enable us to solve problems that we can't solve today with traditional high performance computers. That's fundamentally down to the way that quantum works. So, long story short, I think that quantum will accelerate the AI wave. It will because of some of the things that it will be able to do. But I think it is so significant from a security perspective, from a networking perspective, and from a computing perspective that it will become a wave within its own. Right.
[00:44:39] Speaker B: Wow, that's crazy. A wave within itself. So then Quantum's coming up a lot more now from a media perspective. I'm getting a few more people.
I've got an interview coming up about Quantum. More specifically. I haven't spoken a lot about it, so that's really interesting. The other thing is, would you say there are an adjacent waves happening? So my version of an adjacent wave would be, for example, NFTs and that sort of wave. And then Web3 don't really hear much about that anymore. I think NFTs are completely like obsolete. But Web3 made a quite a large sort of hit in the market and then don't really hear of it anymore. So do you sort of see that these sort of side waves happen and then they sort of die down a bit as well?
[00:45:20] Speaker A: Yeah, I do. Anyone remember the Metaverse? You know, so they, they do. And, and I think that when we were talking at the very beginning, I said, you know, these are sort of the waves from a, an employee working for an intelligent data infrastructure company looking out into the market at what are these kind of big waves that are going to give us huge opportunities ahead of us because they materially offer us opportunity or materially affect the industry that we work in. So my view of kind of what a waves are related to who I am and where I work but you're absolutely right. You know, Blockchain was a wave and Metaverse is a wave and NFTs and Web3. And, you know, I'm sure if you look at it from different perspectives, you could see kind of different waves. But I think some of these things, what did Gartner call it? The. They go from the. So the peak of inflated expectations into a trough of disillusionment and then out onto the plateau of productivity, I think they say. So sometimes we overhype something and then it doesn't sort of deliver against expectations. So we sort of think the wave has died. It's actually just taking longer than we thought. Cloud's a really good example. I remember there was a time back in, I don't know, the early part of the 2000s, maybe 2014, 2015, where every conversation was around Cloud and it was gonna solve world debt and fix world hunger, and it just took a lot longer to materialize into real use cases that people could actually kind of find. It doesn't mean that a wave wasn't gonna form, but it just fell into this trough of disillusionment because people couldn't see the use cases. And we'd so overblown, overhyp and probably over promised on how quickly it was going to happen, how quickly it was going to deliver value. But then we all know what happened with Cloud. It's now a $600 billion business. It's found its use cases, it's found its niche, it's found its opportunity to create value. So now it truly becomes a wave. So again, sometimes a wave can be there ready to form, but it goes off the radar because it doesn't deliver quickly enough. It doesn't mean it goes away, it just means that maybe we didn't get the timing quite right, maybe we were a little bit too early. So maybe that's the way it is with things like Web3. Maybe it's the same with Blockchain. Maybe they were just a little bit too far away from finding those real use cases. But it doesn't necessarily mean they've gone away.
[00:47:45] Speaker B: So lastly, Matt, do you think Quantum will solve world hunger?
[00:47:51] Speaker A: I wish that would be the case. I don't think so. I think it needs an awful lot more smart people and sensible people to get together to solve some of those kind of problems. But I'm a glasses half full person. You can put me in any situation. I will always try and find the bright side of it. I will always try and find the positive side of things. And whilst when I talked about Quantum. I talk about it from the fact that it's going to break encryption, et cetera, et cetera. And the reality is it's also going to do things that will be fundamentally beneficial for society. You know, if we can start to model how cancers change and evolve inside people, because Quantum will enable us to do that, that is phenomenally powerful. If we can model how viruses mutate and evolve in a way that we can't do today, that is an incredible breakthrough. So I don't think it'll solve world hunger, but I do think it will give us technology that will solve some of these really big, challenging problems such as cancer, such as viruses, such as, you know. So yes, I think there is a really interesting positive way that we'll be able to use Quantum that I think will benefit humanity.
[00:49:02] Speaker B: This is kabcast, the Voice of Cyber. Thanks for tuning in. For more industry leading news and thought provoking articles, visit Kpop UBI Media to get access today.
This episode is brought to you by Mercset. Your Smarter Route to Security Talent Mercset's executive search has helped enterprise organizations find the right people from around the world since 2012. Their on demand talent acquisition team helps startups and mid sized businesses scale faster and more efficiently.
Find out more at mercsec.
[00:49:38] Speaker A: Com today.
