 Hey, everyone. Welcome to this CUBE conversation featuring Pepper Data. I'm your host, Lisa Martin. Today, I'm very excited to be joined by Ash Munshi, the CEO of Pepper Data. Ash, thank you so much for joining us on theCUBE today. My pleasure. Thank you for having us. It's been a while since you've been here, so just get the audience a little refresher, a little bit about you, a little bit about Pepper Data and what you guys are doing for customers. Sure. So let me talk about the company. The company has been around for gosh, a decade now. It was founded by two engineers from Yahoo, originally focused on Hadoop and making Hadoop work much better, so providing visibility and also optimization within the Hadoop environment. Since then, the company has evolved, evolved because the Hadoop market has evolved, because big data has evolved and because we see other market opportunities that are there for us from an optimization perspective. Hadoop has gone from on-premise, which large clusters that people were doing to host it on places like AWS. So we have sort of EMR, which is sort of the modern version, if you will, of Hadoop. That's in the cloud, which is growing very nicely as a business for Amazon, where a lot of our on-prem customers are also migrating, including some of our largest customers. And the other place that we see a lot of activity is in the Kubernetes base and in managed Kubernetes, so particularly EKS. And we see the market opportunity not just from a visibility perspective, but from more from an optimization perspective. Our core strength is basically being able to reduce costs by being able to utilize resources much better. We did that for many years on-premise. We now do that on the cloud. We do that with EMR, and we do that also with EKS. And the percentage savings that we can deliver is significant, anywhere from 20% to 40% cost savings, which is quite significant. And we work with some of the largest customers in the world. I think two Fortune 10 companies, several Fortune 100 companies, we are very strong, particularly in the banking sector and in tech. And we're seeing some really good business coming our way. Excellent overview. Thank you for sharing that. When we talk about Big Data, I was just looking at a recent blog post actually on the Pepper Data website and it was published just like a few weeks ago. And it said, IT spend on Big Data systems in 2023 is forecasted to be $222 billion that comes from Statista. Now, we know companies today rely on Big Data to improve decision-making, to deliver better customer experiences, all customers have these expectations, right? But we also know that managing Big Data performance has always been a pretty difficult problem for customers to solve. You talked about Pepper Data and its founding, founded to solve this problem. In the years since 2012, how has Big Data, Hadoop in the cloud really shifted the way that business is being done today? Well, so when Big Data really started, its origins started really within Google and Yahoo, right? So they use this infrastructure to do targeted ads and sort of analytics and real-time analytics to be able to deliver value to their shareholders by being able to sell more ads. That then matured into going into the rest of the world where they said, ah, I can actually do extremely large-scale analytics. I can go beyond my data warehouses and being able to do much better analysis of what's going on. And at the same time, I can do exactly what Yahoo and Google are doing. I can actually give a better customer experience by being able to target it to the customers that come to my website. Initially, the only way you could do this was buying machines that you put into your data center and you had lots of them. So remember, when we're talking about Big Data, we're talking about large distributed systems. So we're not talking about something that is two machines or three machines or whatever. We're talking about hundreds of thousands of machines. So you can imagine that the CapEx cost to that was in the millions, tens of millions and even for the largest customers, hundreds of millions of dollars. And you have many of these systems that were around to being able to do that. That then got to the point where they were delivering tremendous value, but the cost structure was very high. So very few people could actually afford to be able to do that. Income the cloud providers, particularly Amazon. And as Amazon came in, Amazon provided EC2, which was basically, hey, I want to go and do the Hadoop myself on the cloud. And what Amazon realized was lots of people wanted to do that. So they then started offering a service called EMR. And then, which is basically elastic MapReduce, right? Which is Hadoop in the cloud. They started offering that as they started offering that more and more people decided, hey, I don't need to manage it in not only on-prem, but I could actually do this in the cloud. And what that allowed people to do is sort of mere mortals to be able to say, I can afford to do Big Data because I can start with something small and expand it and dynamically expand and contract it. And so therefore I don't have CapEx that I have to worry about. I can do this with OPEX and I can start doing much more analytics. So it became much more widespread to be able to start doing sort of big data and big data analytics. And particularly the analytics is what drove this thing, right? So analyzing my markets, analyzing my sales, analyzing in every way that I could possibly think of that in the old days, I would have had to do a big data warehouse to be able to do. I could now just sort of do queries against these large amounts of data. That became very, very popular to do and it's still continuing to be very popular. I would say that there's still people moving off of on-prem onto EMR and EMR by itself is actually seeing growth, right? So it's not just that, you know, people think Hadoop has disappeared. Well, not really. Hadoop has been replaced by Spark. And so most of the data stacks now are on Spark. Spark is here to stay. Spark is the computational engine that a lot of people use for Big Data. And that is sort of the place that people are doing the building, larger and larger Spark applications. And that's becoming very pervasive through the entire ecosystem. In fact, if you look at somebody like a Databricks, Databricks started as a fundamentally Spark company, they have still the largest Spark developers or committers rather that they're there. Then they built the entire infrastructure on top of that. They're doing more ML and things of that sort. And it turns out that Spark can actually be used for being able to do ML as well. So as ML started taking off, guess what happened? Big Data became broader. It wasn't just analytics anymore. Now you could do ML on top of all this stuff. And ML by definition uses a lot of data. So Big Data became really big data. And now you had two uses. You had both analytics and you had machine learning that you could do. So you could sort of understand what your users were doing, build models and serve those models back out again from a customer action point of view. By the way, exactly what Yahoo and Google did in the entire infrastructure they built over those years. So it's become more and more popular. It's gaining ground with the adoption of ML. And I mean classic ML. And to some extent deep learning as well. It's really the technology that people are betting on. People are betting on it. I was looking at, you mentioned cost. And I want to talk about that because I noticed on the Pepper Data website a bold statement slash your cloud cost by up to 50% immediately. I also saw optimize your cloud price performance by up to three X. Talk about cloud cost optimization and why you've seen it become an imperative in today's market, which is quite dynamic. Yep. So, you know, we had sort of the go-go days between about 2012 to 2021 where nobody cared about cost. Money was free. In fact, it was. The government was printing it and giving it to people. You know, so money supply was large. It was very easy to get loans. There were a ton of startups that were spending money as drunken soldiers basically. And so nobody cared. Everybody was building massive systems to be able to do all of this wonderful stuff. And then we sort of had interest rate hikes and we have a potential recession coming. And all of a sudden in the last probably 12 months especially the last 12 months everybody's talking about, oh my God we've got a cost problem. We've got a cost problem. And the reason is very simple. Capital costs has gone up. You can't borrow money like crazy. And as a result, you have to be careful about what you do. We've been saying this all along, right? You know, we've been saying be careful, be careful of your costs. But now our value proposition is in vogue. We've been doing the 20 to 50% savings for our customers for a decade, right? We have, as I said, some of the largest clients in the world and we are saving them millions of dollars a year by using our software, right? That is even more valuable now in the cloud because in the cloud, two things are happening. One is I can get instances whenever I want. So what do I do? I can expand this elasticity in the cloud, right? That's the reason why people go to the cloud. Well, what happens with the elasticity is you can also become a drunken soldier in elasticity. You can say, oh, you know what? I got five machines to go do this. Oh, don't worry about it. Flex to 500 when you need to. Well, when you start flexing to 500 nodes, all of a sudden costs also start adding up and you've got the same type of issue again. And in this cost sensitive environment, everybody's asking, how do I shrink my costs? Where are my costs? Where am I spending money that I shouldn't be spending money? And there are two ways to solve that problem. One is just analyze the stuff and say, here's a sheet. Here's where all your costs are. And now you manually go in and try to trim the costs. Certainly a perfectly reasonable way to do that. You do that in the house, right? For your own particular balance sheet at your house. But these are dynamic things. And when dynamic things happen and when they scale up and down, what you want to make sure of is they don't scale too fast. They don't scale too slowly, right? That's where you can eat up a lot of costs as well. So we monitor that on a continuous basis. And then we make sure that you use only the resources that are required to do the computation you need. As a result, our customers see actual tangible benefits. So when we say instantly, we do mean instantly. We can take a customer's, we have customers where they take their curve, which is the cost uses report out of AWS. We tell them, take a look at that, install paper data, turn us on, look at your cost use report 24 hours later and you'll see a reduction in cost. No change, no nothing, turn us on. That's all that's required to go do that. And I will tell you that we see savings again between 20 and 50% in simple experiment, right? And so the value proposition is simple. Turn us on, see if you get cost savings. If you do pay us, if you don't, that's okay. We'll go on to the next customer. The good news is we're seeing a lot of connect. We're seeing a lot of people not using the resources efficiently. And as a result, we have a great opportunity in front of us. So we talked only about big data, but the other place that everybody's moving is to Kubernetes. And Kubernetes is a place where you can sort of unify both big data as well as microservices. Well, it turns out microservices are also wasteful of the type of resources as well. So we have a unified platform that allows us to have savings in microservices as well as on the batch or the big data side. So we have a unified solution and we're unique in that aspect, right? We're unique in being able to deliver a total solution for EKS and provide the same kind of savings across an entire gamut. That's kind of where we are. So you really hit on cost and the clear value proposition that pepper data has that it's really substantiated besides costs and all the things that you can do there in the cloud in Kubernetes to help customers really slash that. What are some of the other trends that you're noticing? Well, you know, obviously overall, you can't have any conversation about trends without talking about chat GPT, right? Everything is chat GPT. We're all going to lose our jobs. Chat GPT is going to do everything under the sun. The reality if it is that if you look at models like chat GPT or any large scale models that people are building, they're all data driven. Fundamentally, there's got to be a massive ingest of data in order to do that. So when you think about training a model, before you train the model, you have to ingest lots of data. You have to condition the data. You have to clean the data. And, you know, it's kind of the old fashioned work of digging up the dirt before you find the gold and you got to have lots of that dirt. All of that, all those processes that feed the data engine that eventually becomes a training data set or a chat GPT, that's all done with the kinds of applications we support. And that's fundamentally where we can add value. This, you know, if you chunk it up, you got to get the data, condition it. Then you got to build the models and then you got to serve the models. Well, that entire front end, which happens on a continuous basis, is where we can add a lot of value. So the cost optimization value prop, clear as day. You must have some favorite customer stories that really articulate that value prop and others that pepper data is delivering. Share with me if you will some stories that you think really demonstrate what pepper data is delivering. Absolutely. You know, as I told you, we have a number of Fortune 100 customers and we have some Fortune 10 customers as well. One of our customers, you know, installed us on EMR. They looked at their cost report and I think it was a Tuesday or something and then turned us on, came back, looked at it on the following day, got the updated cost report and low and behold, 24% savings. That was without doing any work whatsoever on their part. Needless to say that was kind of nice, particularly because they spend a lot of money. Think tens of millions at minimum on EMR alone. So that was a nice savings that we had. We also have a very large retailer as a customer and they use us continuously, both on-prem as well as on other clouds to be able to get savings. And we're also delivering something on the order of 20 to 50% savings depending on the type of workload they've got. The third customer, which is very interesting is a technology company that started off by saying, well, I'm not sure that you're going to save us a lot of money, we'll try you. Well, they tried us 90 days later, they bought more capacity. 90 days later, they bought more. I think they've bought five times, they've now bought more capacity throughout this entire time period and it's been great because they're turning into a really large customer for us. In and of when I say large customers for us, I'm not talking about customers that are, you know, $10,000 or $100,000. We're talking about millions of dollars from each of these customers. So we have significant revenue and the only way we can justify significant revenue like that is by having significant savings for them, right? Because effectively our pricing is a percentage of what they save. So imagine we're saving millions and we're getting millions of dollars in payment so you can imagine that we're saving them tens of millions of dollars on the other side. That's the value proposition prices growing very nicely. One thing I would like to say is that, you know, with our collaboration with AWS, it's been wonderful. What's great about AWS is that, Amazon in particular is the laser sharp focus on customers. You know, it's unusual for a company to basically say, hey, guess what? I'm going to take you into a customer to save them money on my platform. AWS is doing that. They're taking us into customers where the customers have a cost problem. They're taking us to be able to go and help them save money. It's been joint marketing, joint selling. It's been a really tremendous pleasure to go in and do calls with them because it's a joint desire to make our customers extremely happy. It's been fun. And I hope that the relationship continues in the long run. Certainly customer reception has been very, very positive. We know, well, the AWS customer obsession story. It sounds like that's very symbiotic, Ash, with what you and the PepperData team are delivering. We appreciate you being on the program. You must have fun talking to customers and prospects, showing them the massive impact that PepperData can make in such a short time period. You must be kind of fun every day, I imagine, yeah? It is fun every day. You know, as I said, I'm an old guy and I gotta have a reason to get up every day. It's really nice to see a simple value proposition that somebody can look at and say, I get it. And you know, it's not some airy-fairy thing that a lot of tech companies sell. It's real savings, real money, a real value proposition. Yep, all real, Ash Munchy, CEO of PepperData. Thank you so much for coming on theCUBE as part of this CUBE conversation. We appreciate it. My pleasure and thank you for having me. Our pleasure. And we wanna tell you, keep it right here for more action on theCUBE, your leader in hybrid tech event coverage.