 from Burlingame, California. It's theCUBE. Covering Sumo Logic Illuminate 2019. Brought to you by Sumo Logic. Hey, welcome back everybody. Jeff Frick here with theCUBE. We're at Sumo Logic Illuminate at the Hyatt Regency San Francisco Airport. Our second year here, about a thousand people. Third year of the conference, a really good vibe. And you know, I think it's one of these cases where the market has really come to Sumo Logic in terms of data and data monitoring. And there's so many applications in our business and security and operations. We're excited to have our next guest. He is Lior Beklavich. He's an architect at Informatica. Lior, great to see you. Great to see you too. Absolutely. So you said you've been coming to this for a couple of years. Just kind of general impressions as it's grown. Sure, it's my third year. It's grown very nicely. Always exciting. I think there's a very nice vibe to this conference. I always learn new things. So we've been with Sumo for more than four years now at Informatica. And excited as always. And we've been covering the Informatica show I think all that we looked at since 2015. So we've been doing a lot of work. And you guys are right in the heart of this whole data thing and you've been part of the kind of migration from pretty much pure on-prem to cloud, has rushed to public cloud and then now kind of this hybrid model. So as you kind of look at the data perspective, you know, what's kind of your take as this thing has evolved over the last several years? Sure. So we have been around for 26 years. I think building a lot of on-prem data platforms for being the enterprise cloud data management that Informatica sells with basically getting your data inside or outside the organization from clouds, on-prem or whatever integration pattern you have. And we decided four or five years ago to be a cloud-first company and migrated most of our products to be on cloud to provide them as a service. And for us, it was a huge journey. We needed to take some offering that we had in the cloud, some products and really revamp and building a new microservice architecture and then slowly migrate all the customers. We took us over a year to make that. We currently run on all three cloud providers and really using sumo and monitoring tools to really understand the impact that we have on our customers during this migration. It was a very successful, they hardly noticed that we changed. Only the nice UI, but they hardly noticed the problems that we really changed a lot of things. What are some of the things you learned in that process that you can apply now with just some of your customers in terms of data migration and operating a cloud situation versus a traditional data center? Sure, so I would definitely highlight the need to be able to roll back and the need to always keep really good monitoring it and understanding how the end users getting impacted. So we really kept that in mind. Everything we do, try to always do it side by side. Then when we migrate, we're really sure that it is successful and there's no impact on the customer. So I think that's definitely too harshly monitor everything and be able to roll back when you need to. Because you will need to at some point. But the rollback is funny because it used to be you had, the release cadence was significantly slower than now and now you've got all these kind of micro pushes that are going out multiple times a day. So how does that impact, kind of keeping that safety net and that rollback safety net? So it's interesting. So we actually don't deploy that many times today. We keep the, where we can really impact the customer. So we deploy the things that are not customer urgent impacting production more, but still the really heavy productions of the truly impact the customer. We try to minimize that and make it a very customer aware. Okay, so basically they choose their own windows of maintenance and all of that. But our customers again, hospitals, all kind of very important to them. We are in charge of the data pipeline in those places. So we don't want to just push whatever we can. We really cannot take that. Even the rollback of 1% of the customers, it can be very bad. So we're a bit of more conservative models of deployments. But it actually means we put a lot of efforts in our monitoring, what is going on during those deployments. All right, so what are the big trends that happened? I mean, containers have been around for a while, but we really saw kind of the rise of containers in terms of the popular consciousness with Docker, couple, three, four years ago, and then a couple of years ago, right, that the Kubernetes coming in for the orchestration. From your point of view, how have those things impacted your world and how you do your job and take care of your customers? Sure, so for us Kubernetes is really a great opportunity to standardize the way that we deploy across different products. So we have our platform, but we have also different products, different people across the globe. We're a very multi-globe organization. And to get a standard like Kubernetes to help us standardize, to get more releases, more stable environments, that really solves a lot of problems because we had this migration that they told about, really left us with a lot of clusters across the globe, different time zones, was really hard to standardize on the pipelines and to deploy to really minimize the problems that we give to the end user at the end. So we really took that opportunity to use Kubernetes, to use containers to minimize the difference it has from the developer machine, all the way to production to automate the most as we can. So Kubernetes really is excelling in this. So that's where we really took those containers apart. So today we are in migrating, so not all of that, but we truly see the benefits of standardization of immutable infrastructure as the key components for us. It's just so great, because you have such kind of a longitudinal point of view that the company's been at it for a while and you've been at the company for a while. So another topic I'd love to get your thought is just kind of this exponential explosion of data. I mean, you'd be curious not to know the numbers, but kind of the scale of data in which you guys are dealing with for your customers and how that has changed over the last several years before you even really factor in IOT and this next kind of machine to machine explosion. Right, so we definitely see that explosion of data. And it's not just the explosion, it's also the different types and where data has been on-prem, now moving to cloud, where people want to run of all those workloads, has of course a lot of impact for us as well, need to support all the cloud providers when we used to do a lot of Hadoop on-prem, right? It's all changed now to all the cloud providers. The data is there, so the data move, data locality is a big thing. Now we need to run all those things on the cloud. So I don't remember the exact numbers. I guess we're doing some kind of a 2.5 billion transactions a month, like number of records that we serve. That being, we usually just see more workloads, more people, more use cases for onboarding more data from cloud applications, right? So the data became more dispersed, not just more data, but sources has become like everybody needs to integrate Salesforce or Workday with their on-prem that gives unique opportunities for this kind of data. Well, and it's funny you talk about the workloads because it always used to be, do you bring the workload to the data or the data to the workload and a knock on the cloud is that you got to get all that data into the cloud and pay for the transport of the data and there's data gravity. That said, once you have it in a central location like that, the opportunity to put applications against that data is much, much higher than if you're bringing the data to the application. You see in that, how are customers taking advantage of that opportunity? So for sure we saw, the data does move to the cloud. The thing that we, when we started Informatica Cloud 10 years ago, our entire model was hybrid, so you can still run on-prem because the data was on-prem. And since then our hybrid model that you can still run both on-prem and on cloud, you can see the change, right? You can see more of our agents. We have an agent-based architecture to really being deployed much more on EC2s, on AKS on whatever to run those workloads in the cloud. Right. But I would imagine that the number of workloads applied to each dataset now have increased significantly because now it's in that central repository. Yes, definitely you can see those data lakes being built and mostly in the cloud. And that does give unique opportunity. Right. So just get your perspective after a couple of days here and I have been here for a couple of days, we're just getting started at the show. You know, what does Sumo Logic bring to you and your team? What does it enable you to do that you couldn't otherwise do? Why are you happy to be a customer of Sumo? Sure, so far and foremost that's the democratization of data. I really like to say that internally. To be able in a multi, in an organization that's spread across the globe, really sharing insights based on data, it's very important, right? And when you have many R&D centers that can just send this Sumo query, send the data and show people what they mean, save so much time. And so we use it across, we use the customer success, product management to understand feature being used, SREs, developers, all of those really can communicate based on data, right? And we can, in this microservice role, you cannot do it without that, you cannot do it without linking because the different products that we onboard on the platform will not be able to communicate effectively without that. That's very important. And giving that landing pages dashboard templates for onboarded services to have this kind of a standard of how to monitor, how to operate, that's very important for us. That's great. I'm sorry. And the key place that we brought Sumo in is basically for incident management. So how to understand when something doesn't work, just to try to understand the place radius, which products are impacted, we have a variety of products. So just in minutes, we minimize that thing for hours to minutes, trying to understand what exactly is going on, who's impacted, to update the customer on the trust side of that. Yeah, that's the- But I love the part you talked about the democratization because again, I talk about it all the time, I'll talk about it again, but to drive innovation in a company, I think such a key piece of it is to enable more people to have more information and the tools to manipulate that information. So they see opportunities to make improvements here and otherwise, and it sounds like you guys are really using it for that. Definitely, we- In this case. You know, when you get some people that you never knew that even though we have customer support guys that did some crazy dashboards, that we had no idea it's possible even, and they're really getting chance to work with customers better to really tell the customer, oh, you just did that, and then maybe you'll try this option, we felt that even communicating and really minimize the time it takes for them to understand what's going on, that has been really impactful. With no call to IT to help. It was never being done, right? So we wanted to allow devs and ops to operate, and all of a sudden you're getting customer support, without even telling them. So, that has been- Good. Well, Lior, thanks for sharing your story and really appreciate you taking the time. Thank you. All right, he's Lior, I'm Jeff. You're watching theCUBE. We are at Sumo Logic Illuminate 2019. Thanks for watching. See you next time.