 From the CUBE studios in Palo Alto in Boston, connecting with thought leaders all around the world, this is a CUBE conversation. Hi, but this is Dave Vellante and welcome to this CXO series. As you know, I've been running this series discussing major trends with CXOs, how they've navigated through the pandemic. And we've got some good news and some bad news today and Ed Walsh is here to talk about that. And how you doing? Great to see you. Great seeing you. Thank you for having me on. I really appreciate it. So the bad news is Ed Walsh is leaving IBM as the head of the storage division. But the good news is he's joining a new startup as CEO and we're going to talk about that. But Ed, always a pleasure to have you. You quite a run at IBM. You really have done a great job there. So let's start there if we can before we get into the other part of the news. So you give us the update. You're coming off another strong quarter for the storage business. And I would say, listen, they're sweet. Hard to believe, but to be honest, we're leaving them in a really good position where they have sustainable growth. So they're actually IBM storage in a very good position. I think you're seeing it in the numbers as well. So yeah, listen, I think the team, I'm very proud of what they were able to pull off. Four years ago, they kind of brought me in, hey, can we get IBM storage back to leadership? They were kind of on their heels, not quite growing or not growing, but falling back in market share. Kind of a distant third place finisher. And basically through real innovation that mattered to clients, which that's a big deal. It's the right innovation that matters to the clients. We really were able to dramatically grow, grow all different four segments of the portfolio, but also get things like profitability growing, but also NPS growing. It really allowed us to go into a sustainable model. And it's really about the team. You've heard of talking about team all the time, which is you get a good team and they really nail great client experiences and they take the right offerings and go to market and merge it. And I'll tell you, I'm very proud of what the IBM team put together. And I'm still the number one fan and insider outside IBM. So it might be bittersweet, but I actually think they're ready for quite some growth. You know, Ed, when you came on theCUBE, right after you had joined IBM, a lot of people were saying, oh, Ed Walsh joined an IBM storage division to sell the division. And I asked you on the Cube, you're here to sell division. And you said, you know, absolutely not. So it's always seemed to me, well, hey, it's a good business, good cash flow business, got a big customer base. So why would IBM sell it? Never really made sense to me. No, I think it's a integral to what IBM does. I think it plays to their client base in a big way. And under my leadership, really we got more aligned with what IBM is doing from the big IBM, right? What we're doing around Red Hat, hybrid multicloud and what we're doing on AI. Those are big focuses of the storage portfolio. So listen, I think IBM is a company is in a position where they're really innovating and driving and really customer centric. And I think IBM storage is benefiting from that and vice versa. I think it's a good match. So one other thing I want to bring up before we move on. So you had said, we're seeing a number. So I want to bring up a chart here. As you know, we've been using a lot of data and sharing data reporting from our partner, ETR Enterprise Technology Research. They do quarterly surveys. They have a very tight methodology. It's similar to NPS, but it's a net score, we call it methodology. And every quarter they go out and what we're showing here is the results from the last three quarter specific to IBM storage and IBM's net score in storage. And net score is essentially we ask people, are you spending more, are you spending less? We subtract the less from the more and that's the net score. And you can see when you go back to the October 19 survey, you know, low single digits and then it dipped in the April survey, which was the height of the pandemic. So this was, this is forward looking. So at the height of the lockdown, people were saying, well, maybe I'm going to, you know, hold off on budgets, but then now look at the July survey, you know, huge, huge uptick. And I think this is testament to a couple of things. One is, as you mentioned the team, but the other is you guys have done a good job of taking R and D, building a product pipeline and getting it into the field. And I think that shows up in the numbers. And that was really one of the hallmarks of your leadership. Yeah, I mean, the innovation at IBM is there's almost an embarrassment or riches inside. It's how do you get in the pipeline? We went from typically about four, four year, four and a half year cycles, now to two year cycle, product cycle. So we're able to innovate and bring it to market much quicker. And I listen, I think that's what clients are looking for. Yeah, so I mean, you brought a startup mentality to the division. And of course now, because you're a startup guy, let's face it, now you're going back to the startup world. So the other part of the news is Ed Walsh is joining Chaos Search as the CEO of Chaos Search is a local Boston company. They're focused on log analytics, but more we're going to talk about that. So first of all, congratulations and tell us about your decision why Chaos Search and where you're at there. Yeah, listen, you couldn't tell them from the way I describe IBM. I mean, it was a hard decision to leave IBM, but it was a very, very easy decision to go to Chaos, right? So I knew the founder, I knew what he was working on for the last seven years, right? Last five years as a company, but and I was just blown away at their fundamental innovation and how they're really driving like how to get insights at scale from your data, you know, your data lake in the cloud, but also, and also it's an and statement slash cost dramatically. And they make it so simple, simply put your data in your S3 or really cloud object storage, but right now it's, you know, Amazon, they'll go to the rest of the clouds, but just put your data in S3 and what we'll do is we'll index it, give you API so you can search it and query it. And it literally brings a way to do at scale data analytics and also log analytics on everything you just put into S3 basically bucket. It makes it very simple. And because they're really fundamental, we can go through it fundamental hard technology of the data layer, but they kept all the API. So you're using your normal tools that we did for Elastic Search APIs. You want to do Grafana, you want to do Kibana or you want to do SQL or you want to do, you know, use Looker, Tableau, all those work, which is that's a part of it. It's really revolutionary what they're doing as far as the value prop and we can explain it, but also they made it evolution. It's very easy for clients to go just run in parallel and then they basically turn off what they currently have running. So, you know, data lakes really, the term became popular during the sort of early big data sort of Hadoop era. And, you know, Hadoop obviously brought a lot of innovation, you know, leave the data where it is, bring the compute to the data, you know, really launched the big data initiative, but it was very complicated. You had, you know, map reduce and then elastic map reduce in the cloud and it really was a big batch job where storage is really kind of, you know, the second class citizen, if you will. There wasn't a lot of real time stuff going on and then, you know, Spark comes in and still this very complicated situation. So it sounds like chaos search is really attacking that problem and the first use case that's really going after is log analytics. But explain that a little bit more, please. Yeah, so listen, they finally went after it with this, you know, it's called a data lake engine for scalable and we'll say log analytics firstly. It was the first use case that go after it. But basically what they allow us for log analytics, people, everyone does it and everyone's kind of getting to scale with it, right? But if you ask your IT department, you know, are you having a challenge with scale or cost or retention levels, but also management of what they're doing on log analytics or security log analytics or all this machine data they're collecting, the answer would be absolutely no, it's a nightmare. It starts easy and it becomes a big, very costly application for environments. And what chaos does is because they deal with a real issue, which is the data layer, but keep the APIs on the top end so people easily use the data insights at scale. What they're able to do is very simply run in parallel and we'll save you 80% of your cost, but also get better data retention. Because there's typically a trade-off, clients basically have a trade-off or it gets really expensive, it gets to scale, so I should just retain less. We have clients that went from 90 day retention on security logs to literally four and five days. If they didn't catch it in that time, it was too late. Now what they're able to do is they're able to go to our solution, not change what they're doing on applications because we're using the same APIs, but literally save 80%. And this is millions and tens of millions of dollars of savings, but also basically get 90 day retention. It's really limitless. Whatever you put into your S3 bucket, we're gonna give you access to. So that alone shows you that it's just, it's literally revolution that CFO wins because they save money, the IT department wins because they don't have to wrestle with this data technology that wasn't really built. It was really built 30 years ago. It wasn't built for this volume and avaracity of data coming in. And then the data analytics guys, hey, I keep my tool set, but I get all the retention I want. No one's limiting me anymore. So it's kind of an easy win-win and it makes it really easy for clients to have this really big benefit for them. And dramatic cost savings, but also you get the scale, which really means a lot in security log analytics or anything else. So let's dig into that a little bit. So cloud object storage has kind of become the facto bucket, if you will. Everybody wants it because it's simple. It's a get put kind of paradigm and it's cheap, but it's also got performance issues. So people will throw cash at the problem. They'll have to move data around. So is that the problem that you're solving? Is it a performance problem? Is it a cost problem or both? And explain that a little bit. Yeah, so it's all over the, so basically if people were building a data lake, they would like to just put all their data in one very cost-effective, scalable, resilient environment. And that is cloud object storage or S3 or every cloud has around, right? You can do also on-prem. Everyone would love to do that and then literally get their insights out of it, but they want to go after it with their tools. Is it search or is it SQL? They want to go after their own tools. That's the vision everyone wants, but what everyone does now is because, and this is where the core special sauce, what Chaos Search provides, is we built from the ground up, the database, the indexing technology, the database technology, how to actually make your cloud object storage a database. We don't move it somewhere. We don't cache it. You put it inside the bucket. We literally make the cloud object storage the database. And then around it, we basically built a chaos fabric that allows you to spin up compute nodes to go after the data in different ways. We truly have separated the data for the compute, but also if a worker nodes, the beauty of like container type of technology, a worker nodes goes away, nothing happens. It's not like what you do on prem and all of a sudden you have to rebuild clusters. So by fundamentally solving that data layer, but really what was interesting is they just publish APIs. You mentioned put and get. So the APIs you use in cloud object storage is a put and get. Imagine we just added to that API, your search API from Elastic or your SQL interface. It's just all we're doing is extending. You put it in the bucket, we'll extend your ability to get after it. Really it's an API company, but it's a hard tech of putting that data layer together. So you have cost effectiveness and scale simultaneously. But we can ask for, for instance, log analytics. We don't cache, nothing's on SSD, nothing's on local storage. And we're as fast as you're running Elastic search on SSDs. So we've solved the performance and scale issues simultaneously. And that's really the core fundamental technology. And you do that with math, with algorithms, with machine learning, what's the secret sauce? Yeah, and we should really have, I'll tell you, my founder just has the right interesting way of looking at problems. And he really looked at this differently and went after, how do you make both going after data? He really did it in a different way and really a modern way. And the reason it differentiates itself is he built from the ground up to do this on object storage, where basically everyone else is using 30-year-old technology, right? So even really new up-and-coming companies, they're using Tableau, Looker or Snowflake would be another good example. They're not changing how the data's stored. They always have to move at ETL at somewhere to go after it. We avoid all that. In fact, we're probably a pretty good ecosystem players for all those partners as we go forward. So you're talking about Tom Hazel, your founder and CTO and he's brought in a team and they've been working on this for a while. What's his background? Large telecom, building out God boxes. So he's always been in the database space. I can't do his, in my first day of the job, I can't do justice to his deep technology. There's a really good white paper on our website that does that pretty well. But literally the patent technology is a chaos index, which is a database. It makes your object storage the database and then it's really the chaos fabric that puts around it in the chaos refinery that gives you virtual views. But that's one solution. And if you, like for your log analytics, you come in, log in and you get all the tools you used to, but underneath the covers, we're just saving about 80% in overall cost, but also almost limitless retention. We see people going from, Lily they've been reduced the number of logs they're keeping because of cost and complexity and scale down to literally a very small amount and going right back at 90 days. You could do longer, but that's what we see most people go into when they go to our service. Let's talk about the market. I mean, as a startup person, you always look for large markets. Obviously you got to have good tech, a great team and you want large markets. So the space that you're in, I mean, I would think it started early days and kind of the decision support sort of morphed into the data warehouse. You mentioned ETL, that's kind of part of it, business intelligence, it's sort of all in there. If you look at the EDW market, it's probably around 18 to 20 billion. Small slice of that is data lakes, maybe a billion or a billion plus. And then you got this sort of BI layer on top. You mentioned a lot of those, you got ETL. You probably get up into the 30, 35 billion just sort of off the top of my head and from my historical experience and looking at these markets. But I have to say Ed, these markets have traditionally failed to live up to the expectations. Things like 360 degree views of the customer, real time analytics, delivering insights and self-service to the business. Those are promises that these industries made and they ended up being cumbersome, slow, maybe requiring real experts, requiring a lot of infrastructure. The cloud is changing that. Is that right? Is that the way to look at the market that you're going after? You're a player inside of that very large TAM. Yeah, I think we're a key fundamental component underneath that whole ecosystem. And yes, you're seeing us build a full stack solution for log analytics because there's a really good way to prove just how game-changing the technology is, but also how we're publishing APIs and it's seamless for how you're using log analytics. Same thing can be applied as we go across the SQL and different BI and analytic type of platforms. So it's exactly how we're looking at the market and it's those players that are all struggling with the same thing. How do they add more value to clients? It's a big cost game, right? So if I can literally make you underline how you store your data in an exit, literally, you know, 80% more cost effective, that's a big deal or simultaneously save you 80% and give you much longer retention. Those two things are typically, you know, literally a trade-off you have to go through and we don't have to do that. That's what really makes this kind of an underlying core technology. And really, I look at log analytics as really the first application set, but if you have any log analytics issues, you know, if you talk to your teams and find out, you know, scale, cost, management issues, it's a pretty, we make it very easy, just run in parallel, we'll do a POC and you'll see how easy it is and you can just save 80%, which is, you know, 80% and better retention is really the value proposition you see at scale, right? So this is day zero for you. Give us the 100-day plan, what do you want to accomplish? Where are you going to focus your priorities? I mean, obviously the company's been started, it's well-funded, but where are you going to focus in the next 100 days? No, I think it's that. It's building out where are we taking the next, you know, there's a lot of things we could do. The degrees of freedom as far as where we'd go with this technology is pretty wide. You're going to see us be the best log analytic company there. We're getting, you know, really a dramatic, you saw an announcement, best quarter ever last quarter and you're seeing this nice as-a-service ramp. You're going to see us go to VPC, so you can do as-a-service with us, but now we can put this exact same thing in your own virtual private data center. You're going to see us go to Google Azure and also IBM Cloud and the really clients are driving this. It's not us driving it, but you're going to see actually the client, so we're going to Google because we had a couple financial institutions that are saying they're driving us to go do exactly that. So it's more really working with our client sets and making sure we got the right roadmap to support what they're trying to do. And then the ecosystem is another play. How to, you know, my core technology is not necessarily competitive with anyone else. No one else is doing this. They're just kind of, hey, move it here. I'll put it on this, you know, foundational DB or they'll put it on a Presto environment. They're not really worried about the bottom line economics, which is really that's the value prop and that's the hard tech and patented technology that we bring to this ecosystem. Well, people are definitely worried about their cloud bills. You know, the CFO saying, whoa, because it's so easy to spin up, you know, instances in the cloud. And so, so, Ed, it really looks like you're, you're going after a real problem. You got some great tech behind you. And of course we love the fact that it's another Boston based company that you're joining because it's more Boston based startups. The better for us here at the East Coast Cube. So, Ed, give us your final thoughts. So, you know, what should we look for? I'm sure we're going to be in touch and congratulations. No, hey, thank you for the time. I'm really excited about this. I really just think it's fundamental technology that allows us to get the most out of everything you're doing around analytics in the cloud. You know, if you look at a data lake model, I think that's our philosophy and we're going to drive it pretty aggressively. And I think it's a good fundamental innovation for this space. And that's the type of tech that, you know, I like. And I think we can also, you know, do a lot of partnering across the ecosystems to make it work for a lot of different people. So anyway, so I guess thank you very much for the time. I appreciate it. Yeah, well, thanks for coming on theCUBE and best of luck. I'm sure we're going to be learning a lot more and hearing a lot more about Chaos Search. Ed Walsh, this is Dave Vellante. Thanking you for watching everybody and we'll see you next time on theCUBE.