 Live, from Las Vegas, it's theCUBE. Covering Informatica World 2019. Brought to you by Informatica. Hey, welcome back everyone here. Live in Las Vegas, for theCUBE, for Informatica World 2019. I'm John Furrier, co-host of theCUBE. We've got two great guests here from Snowflake. We've got Tarek Diwak, who's the Director of Technology Alliance at Snowflake, and Rick Tam Daniels, Vice President of Strategic Ecosystems and Technology at Informatica. Welcome back to theCUBE. Good to see you guys. Good to see you as well. Hey, thanks for coming on. Snowflake, congratulations, you guys are doing really well. Thank you. Big growth, new CEO, Frank Slutman. Informatica, the data, ZAR, Neutral Third Party, Switzerland, Cloud, you got Switzerland. What's the relationship, explain. Well, I think it's funny that comment comes up a fair amount, and I look at it this way. It's not so much that, you know, with Switzerland, what we're focused on though, is where customers are choosing to go in their journey, we want to provide them the best experience possible, right? So we end up going very deep in our strategic ecosystems, and Snowflake's one of those partners that we've seen tremendous growth with, and customers are adopting. So, very excited about the partnership. Talk about your relationship with Informatica, why you're here, what's the story? Yeah, definitely. So, Snowflake, we put customers first, right? And as Rick mentioned, it's all about having a diverse ecosystem, and the enterprise, Informatica's a leader, when you look at where customers are going with data, right, they, obviously data integration's key, data quality is key, data governance, all the areas that Informatica has been the best to read in, it just makes sense for, continue to make traction in these enterprise customers. Can you explain the business model of Snowflake, what you guys do, quick one minute? Sure, so Snowflake's a data warehouse solution built from the ground up for the cloud. Why that distinction's important is because we're the only data warehouse born in the cloud, if you look at how the other solutions you're doing today, they're taking an architecture, an architecture created a decade ago for a non-premise world, and they're just shifting to the cloud. And the challenge that you have there is that you can't take full advantage of things like instant and infinite resources, both compute and storage, right? Independent scaling of compute and storage, elasticity, right, the ability to scale up and down and out with the click of a button, and then even being able to support massive concurrency, things like loading data, at the same time that you're querying data. This is what Snowflake was built for. How about data sets from other people? That's one of the benefits of having cloud data in the cloud. Correct, so our architecture's key, that's the key to our business and our product, and what we've done is we've separated compute and storage, and we've become a centralized database, and what we've found by creating additional views, you can actually share your data with yourself, you can share it with other customers. We've created this concept of data sharing. Data sharing's been around for decades, but it's been very painful. What we've done is created an online, performance secure way for customers to share the data. Rick, this is really highlights of value proposition for Matica. I always say, beauty in data is in the eye of the beholder, depending on where you're sitting in from. You could be on-premises, you have legacy, you could be born in the cloud, and taking advantage of all that cloud stuff. Graham Thompson was on earlier, he said, hey, if you've got data in the cloud, why move it on-premise? So there should be a choice of what's best, and that's where you guys come in. What specifically are you guys tying together with data where it's in the cloud, and make customers want to choose to have, for compliance reasons, a variety of other reasons, on-prem or under the location? Well, I think one of the big things about cloud data warehouses in particular, it's not all things being equal with the on-premise world, right? The level of agility you get with the snowflake, where it's infinite scale out, up in a few minutes, that empowers so much transformation in the organization, that's why it's so compelling, and so many folks are adopting it, and so we're helping customers on that journey though, because they've got a very complex data environment, and they've got to, first of all, understand how does this all put together to be able to start modernizing and moving to the cloud? I'm sorry, if I ask the question, where should a customer store their data on the cloud or on-premise, I know where you'll come in on that, it's cloud all the way, because that's what you do. But this is something that architects and the enterprise have been feeling with, because they do have legacy stuff. So, and we've seen with the SaaS business models, data has been really key for their success, because it gives them risk-taking, or I'm saying risk-taking, meaning they can do things, A-B testing to whatever, test certain features on certain users, basically use the data, basically, to create value. And then the upside of taking that risk is reward. You get more revenue, hockey stick growth, and the numbers are pretty clear. Enterprises want that, but they're not really set up for it. How do they get there? The best part with the SaaS model is customers can de-risk by putting some of their data, for instance, Snowflake, right? We work across AWS and Azure. So, customers that maybe aren't all in yet on either cloud provider can start using Snowflake, and put data into Snowflake, and test it out, test out the performance and the security of cloud. And if, for whatever reason, it doesn't work out, they haven't risked very much, if anything. And if it does work out, then they've got a great proving round for that. So, the SaaS opens up a lot of possibilities for enterprise customers. Jay, that covers, I brought this up with Graham, and he's from Scotland, so I understand his perspective. I'm from Silicon Valley, so I took my perspective. I said, when I hear regulation, I see anti-innovation, right? Like, when I hear governments coming involved, putting regulation on things, we're seeing a very active regulatory environment on tech companies around data. GDPR, one year anniversary, this is a real issue. How do you turn that regulatory constraints around data, because what it means is more complexity around how to deal with the data. How do you turn that into an advantage? Obviously, software abstraction certainly helps from tech, but customers are trying to move faster with cloud. They can do that for all those reasons we talked earlier, but now you've got complexity around regulation. I think, first off, from a data warehouse perspective, we were built with security and compliance in mind from day one, right? So you build in things like encryption, always on encryption. You build things like role-based access controls, things like key management, right? And then, when you think of Informatica within the data pipeline, getting data from sources in and out of Snowflake, then you build additional data quality, data governance tools on top of that. Things like data catalog, right? Where you can now just go discover what data you have out there, what data are you moving into the cloud, and what is the lineage of that data? So, about this migration and movement, because that becomes, people are generally skeptical when they hear migration, like, oh my God, migration. Either they know it's going to cost some money or potentially technical risk. How do you guys handle the migration in a way that's risk-free? Yeah. I'll take that one. So, one of the things that we really put in front of all of our migration approaches for customers is the Enterprise Data Catalog. And using the machine learning capabilities of the catalog to take what is a very complex landscape and make it very understandable and accessible to the business, but then also understand how it's all put together. Where data's coming from, where it's going, who's consuming it, and once you have that view and that clarity of how things are put together, it actually means you can take a use case-based approach to adoption of the cloud and moving data. So, you're actually realizing business value incrementally as you're moving, which I think is really key, right? If you do these massive multi-year projects and it takes a year to get any results, it's not going to fly anymore, right? This is a much more agile world. And so, we're really empowering of that with the intelligence around data. You know, Digital Transformation's got three kind of categories we find when we poll people and do research. You've got the early adopters who have a full team, they're cloud native, they're jamming, they're DevOps, rock stars, they're kicking ass, taking names. Then on the other end of the spectrum, you've got fear, oh my God, I don't really have the talent to do some study it, spec it out, we've got to figure it out. Then you have people who are kind of like, you know, the fast followers, influence kind of like, focused. They tend to break down in the middle of projects. And the sun seems to be the pattern. They get going and they get stuck in the mud. This is a real issue around culture and people. So, I've got to ask you, you know, a lot of these challenges around people and culture is huge skills gap. What is the biggest hiring skills gap that's needed to be filled so that people can be successful, whether they've got a really rock star team or smart team that's just got to reskill up or how do you take a project that's stuck in the mud and rebooted, these are challenges. I think one of the nice things about Informatica is that there's 100,000 folks out there who are familiar with Informatica's approach of implementations. So, by us bringing our technologies and embracing these journeys, we're actually empowering customers to not have to get coders and data scientists. They're using some of those same data engineers, but now they're bringing data to the cloud. And I think along those same lines, we think of practitioners usually, right? I need data scientists, I need more data engineers. I think a valuable asset that's becoming more clear now is to have a new breed of data analysts, right? That understand how to put AI and machine learning together, how to start to grab all of the data that's out there for customers, right? Structured data, semi-structured data, and make sure that they've got a single strategy along how to become data-driven. Give an example of some of the customers you guys are working together with using Snowflake and Informatica. What are they doing? What's some of the use cases? What's some of the applications? Yeah, so I think one of the biggest use cases is data warehouse modernization, right? So you have the existing on-premise data warehouses, and I always like, I talk to customers, think about, well, realistically, when you have a new use case on your on-premise warehouse, how long is it going to take you to actually see your first piece of data? I don't know, a lot of people have extra capacities, kind of hanging around in their warehouse, right? When you think about, they have to make business cases, they have to get new hardware, new licenses. It could take six months to see their first piece of data, so I think it's a tremendous accelerator for them to go to the cloud. So the main thing there is agility. Yes, exactly. Fast time to value. How's business with the Snowflake? What's going on with you guys? What other use cases are you seeing besides the data warehouse, modern data warehouse? Sure, John, I can start with business in general. It's very exciting times at Snowflake right now. Late last year, we got a funding round of $450 million for growth funding, brings our total funding to just over $920 million. Our valuation doubled to $3.9 billion. That puts us in the top 25 highest valued private U.S. tech firms. Like I mentioned before, we tripled the number of employees to over 1,000 across nine countries globally. We're going to expand to 20 or more in the next 12 months. And then in terms of my favorite part. What's been the traction of that? Why the success? What's been the aha moment for customers with Snowflake? Yeah, I think about what customers try to do in their data journey. There are probably three key things. Number one, they want to get access to all their data. And they want to do that in a very fast and economic way. They want to be able to get all the different variety of data that's out there. All the modern data types. Both the structured data, their ERP and CRM systems, things about customers and product and sales transactions. And then all this modern data from web and social, from behavior data, from machine generated data and IoT. But they want to put it all together. They don't want to have different disparate systems to go and process this and try to bring it all back together today. That's been the challenge. Is the complexity and the cost. And what we've done is start to remove those barriers. You know, I love the, well, I love the term now because I hated it when it came out data lake. During the Hadoop days we heard data lake. And then it turned into a data swamp. You start to see that get fixed a little bit. Because what people are afraid of is they're afraid of throwing all this data into a data swamp. They really want to get value out of it. This has been a hard thing in the early days of Hadoop. But it was cool technically to be, you know, putting Hadoop clusters together and standing them up. But then it's like, where's the value? I think the data lake concept, in essence, makes a lot of sense. Because you want to get all your data in one central place so you can ask questions across all the different data types and all the different data sources. The challenge we had was you had the traditional data warehouse, which couldn't support the new data types and the diversity, just the pure volume. And then you had newer, no sequel like systems like Hadoop that could start to address just the sheer mass of data. But they were so complex that you needed an army and you still do need an army. And then there's some limitations around performance and other issues. And so no data projects were making it into production. I think we still have a very small success rate when you think about data projects that actually make the production. This is where, with Snowflake, because we had the luxury to build it from the ground up, we saw the needs of both using a relational SQL database because SQL's still an amazing expressive language. People have invested skill sets and tools. And then be able to support the new semi-structured data types, all within the same system. All within a SaaS model so you can start to remove complexity. It's self-managed. We have a self-managed SaaS offering so customers don't have to worry about all the operational lifting. They can go and get insight into the data. And then because of the cloud, they can take advantage of the elasticity and the scale and pay for what they use. What was the big bet on Snowflake that paid off? You had to kind of hone it down. But the biggest bet, John, was re-architecting a database in scratch because if you look all the other solutions out there that get the fastest time to market is you can take an architecture that's been existing for a decade or so and wrap it around a cloud. And that gets you some benefits of the cloud. For instance, no need for upfront costs and implementing hardware in the data center. You can offload some of the management and some of the maintenance to the cloud providers. But like I mentioned before, you can't scale automatically. You can't take advantage of infinite scale, right? Because these systems were designed in an on-premise world that had a thinking of finite resources. So I think our big bet was, do you create a new architecture? That's a big risk, but luckily it's paid off well. Big risk payoffs. Rick, talk about the ecosystem. You guys have a big partner strategy. You have to. You guys are integrating integration points. I was comparing you guys, not to sound like it's in a bad way, but Slack is going public, so I'll use them as an example. Slack is a software that's cloud-based, but what made them really big besides copying the message board kind of IRC chat, is that they have a huge integration points with all the key players that really fed that in. This is kind of something that in metaphors, not directly to you guys, but you guys are very integration, partner-oriented. How is that playing out? Again, I'm sure there's, I didn't see any strategy change, so I'm still continuing. Give us the update. How's that going? This is a great example. It's no-flake here on theCUBE. This is core of Informatica. Take a minute to explain the strategy. Well, I think the beginning of the journey with any of our ecosystem partners does start with the connectivity layer, but honestly, moving data from point A to point B, that's kind of, that's the tip of the iceberg, right? And so we've really focused on bringing, really addressing all the challenges, the entire data journey. So it's one thing about, first of all, how do we even find the data to bring there? Now once I've found it, can I connect to it? Do I have the access to the data? Can I bring it to the right targets the customer wants to consume? But then once the data is there, is it usable? Is it consumable? Is it clean? If I'm doing customer 360, do I need to get my golden records? Or you mentioned GDPR, our whole data protection focus on trying to create a perimeter between different parts of the enterprise or automatically applying masking, encryption, those sorts of things. So we're really focused on integrating that as tightly as we can and making it seamless for customers to be able to tap into those capabilities when they need them. I mean, feeding data to machine learning and then powering AI is a great example. You don't have the right data at the right time for the machine learning. The AI doesn't work well. And then applications that are going to be using machine learning need to have access to data as fast as possible. Lag really hurts everything. And this is a huge issue. Yeah, I mean, and we're looking at complete acceleration of that whole data discovery phase to build your models and train them. But your point, garbage in, garbage out, right? The old adage is still applicable today. And I think even, but you've got security issues. What happens if your training data includes some sensitive code names that show up in your models all of a sudden, right? There's all these issues, but then you take out those models and operationalize them as well. Again, the inputs need to be clean. Cloud on premise, final word. Get your, get your bull's take on it. Obviously your data where it's in the cloud. For the customers that have an on-premise dynamic, whether it's legacy or whatever, I got to move to the cloud. I'm eventually going to have some cloud and how it's going to look. What do they do? What's the state of the union for dealing with data that's not just in the cloud? Yeah. I guess, go ahead. Sure, I think, again, going back to having a SaaS model, customers can pick specific projects, specific data sets to go and try out, right? Snowflake gives them a perfect example of not even having to directly engage the cloud partner yet, right? They want to see if data can be ingested in the cloud in a very fast, performant way. They want to see if security meets their needs, right? They want to test out all of the different things around management and ease of use. They can do that with Snowflake. Again, at a very low risk way because we are a SaaS platform, we've got a great model on elasticity, customers can pay as they go just to try it out. So for me, when I think of these customers that are stuck there and trying to make a decision, I say, look, try Snowflake. It's a very risk-free way to start to analyze some data sets. And if it works for you, then you've got a proof point of starting to move more and more workloads into the cloud. Rick, digital transformation, what are customers doing? What's the playbook? Yeah, I think the recipe is, one, be laser-focused on value, right? Have your eyes on how I'm going to get value as quickly as I can this transformation. Second thing is, understand what you have. Understand your existing landscape. And third piece is go, like get started because I think the case for the cloud is so compelling for customers. I don't know a single customer that I talked with who is not already on the cloud journey. So it's really though about making sure you get business value as you proceed down that journey. Get the proof points up front. Absolutely. Think smaller steps. Yep, incremental. Jill of the value. Sounds like agility dev outs. Guys, thanks for coming on. Good to see you. It's Cube Coverage here in Las Vegas. I'm John Furrier, your host of the Cube with Rebecca Knight. Two days of wall-to-wall coverage. We'll be back with more after this short break.