 Live from San Diego, California, it's theCUBE. Covering KubeCon and CloudNativeCon, brought to you by Red Hat, the CloudNative Computing Foundation and its ecosystem partners. Welcome back, I'm Stu Miniman, my co-host, John Troyer, and this is theCUBE's fourth year of coverage of KubeCon, CloudNativeCon 2019. We're here in San Diego, and happy to welcome to the program a first-time guest, Amar Abdahallam, who is the head of Cloud Platforms at Fidelity Investments. Of course, Fidelity, we love talking to an end-user, big financial company, your boss was up on the main stage in front of 8,000 people. Just in that room, there's over 12,000 here in person. Fidelity itself, founded in 1946, first computers in 1965. In the last year, you've now got over 500 applications running in the public cloud, and Fidelity also joined the CNCF. So let's start there, Amar, if we would. Just how does Fidelity look at Kubernetes and the CNCF? How does that fit into your company's mission? Absolutely, thank you so much for inviting me here. Innovation in Fidelity is a big part of the process. We're very focused at this time in cloud computing, in machine learning, in AI technology. We had the first financial robo in 2015, I believe. We had the first augmented reality financial advisor who was actually released this year as a prototype. So Bartels at Innovation, we're seeing CNCF and the cloud computing and cloud native is the key for strategy for our innovation. All right, maybe if you could, give us a little bit of the breadth and depth of your team, what they cover, cloud platforms. What does that mean inside of Fidelity? Sure, so Fidelity had over 10,000 of IT, like hundreds and hundreds of DevOps teams, 1,000 of applications, it's globally distributed. It had all kind of workloads that you can imagine, and it's running in very high-regulated environment as well. And that's what we're seeing, that we're always looking for this autonomy between teams and agility and improve time to market and customer experience, and the key for that is cloud native. We're seeing Kubernetes and CNCF and the cloud native technology is like a key player for us when we go like market cloud, hybrid cloud model. Can you talk a little bit about more into that portfolio of technologies? You know, there's a lot of talk about public cloud versus on-prem, and as if one thing is going to, one knife is going to be the only thing you need in your kitchen. Right. So you have a portfolio of platforms, you have a portfolio of destinations and a portfolio of applications. Can you talk a little bit about both what you're using and maybe how you're organized to access and address all those needs? Absolutely. So I think, you know, 2019 I would say it's a year of multi-cloud, hybrid cloud modeling. Right, actually I would say that 2020 is more, going to be more about distributed cloud where you can distribute your workload across multi-cloud providers. We're not there yet. I don't say where anyone is there yet, but at least we should start like somewhere where he has this multi-cloud providing. Distributing the workload itself I mean, it's a journey to move like thousands of applications and thousands of workload and data as well between like on-premises data centers to a public cloud. You will need to move through this journey of hybrid cloud models and be able to move apps like slowly and aggressively to other apps. All right, Amar, I want to dig into what you talked about there, multi-cloud. So, you know, when you talk about multiple clouds, yes, everybody has that. Walk us through a little bit, you know, where you have workloads and how many public cloud you use and like, but I want to set you up with a premise. You know, we've really said, for multi-cloud to really be a reality, the value that you extract should be greater than the sum of its parts. And most of us live through the multi-vendor years, and that wasn't necessarily, you know, happiness and joy when I had to span between those environments. So, you know, how do we make sure that multi-cloud doesn't become the least common denominator or a detriment to what I need to do with my data, my applications, the value that the company has. And that's why we were here. We were here actually in Kubernetes and KubeCon for that reason. That's where we see this abstract layer that guarantee you the portability of moving your application from one cloud provider to another, right? That capability of like, ability to deploy the same workload into multi-cloud. The ability to have the workload itself managed in different, like, characteristics next to like, services that you will find in AWS via Azure, via Google Cloud, via others, right? That's where you need that, you know, flexibility and Kubernetes and cloud native itself. The ability to have the same deployable construct for your application. The ability to have the same ecosystem around that constructor and around that artifact. The ability to move all of that as is from one cloud provider to another cloud provider is big, big key. And that's you can only find with Kubernetes. Are you, can you share like which cloud or cloud you're working on today and what is your roadmap to, you know, do you have a kind of timeline to when that vision becomes reality? I mean, at this moment we're like, we're with a major like cloud provider keys that you guys can name them. All the colors. You're using all of them. All the colors. Okay. And how are you using Kubernetes today? Where are you in that journey? So Kubernetes is mainly, I mean, outside the majority is running, is still running on-premise. We're like very intensively moving to public cloud and the Kubernetes side. At this moment actually we're building an offer inside my team, which is a cloud platform team that offer will guarantee that portability between all the cloud provider. So for the volume team to import our platform, it will be kind of seamless for them. Where is it going to land? Is it going to be landing in AWS or Azure or on-premise? Okay. Joining the CNCF as a member, bring us inside, you know, I understand the journey. Are there any specific goals you have? How do you measure kind of the investment and what you're hoping to both as a company as well as part of the community get out of it? So we have like a big hope right now in open source our project, our little project about multi-clouding. And our focus is mainly about like the higher regulation part, the higher regulated part. We're very focused in compliance and security. And that's where we can, I think we can contribute back to the open source community around that. So, Amar, you talked about, you know, we talked about the platforms here in Kubernetes, but that goes hand in hand with the culture and the upskilling and the organization and the processes. So what intrigued me is you said, well, we put things on Kubernetes on-prem and then, you know, and some things in the cloud, but then we're going to move some of those apps over time, we'll move to other appropriate homes. So that implies that you've changed process and you've changed org maybe to be able to build cloud native apps. And that was actually separate in some cases from being in the public cloud. Is that the, can you talk a little bit about how you've approached, I mean, from the perspective of people who are listening or watching who are, you know, IT admins and wondering how a, you know, a major organization like, you know, your org, you know, gets there. Right. And this is a main challenge. The challenge is not in the technology side itself or the tools. I think the majority is there in the ecosystem at this moment. The challenge is mainly like building the sculpture inside teams. So we're building like, you know, many like star points or COEs across all of our business unit and all of our teams. And again, like to build this culture across 10,000 developer plus, that's a measure. And it's funny because sometimes people go, what's COE is a dirty word, right? Don't do a COE, but you've had multiple COEs distributed across. So it's like nuclear reaction. Like our COE is the first one that will communicate with few COEs. Each one of them will be with other COEs. And that's how that chain will go and expand like quickly. All right. And this is happening at this moment. So, Amar, I have a few friends that this is the first time that they've come. And they go into the keynote or they look at the schedule and they're a bit overwhelmed. They say, you know, it's not just Kubernetes. There's, you know, dozens and dozens of projects. The ecosystem is sprawling. If you could give us a little walkthrough as to the projects you're using. Any key partners that you're allowed to talk about that are useful in helping you to achieve your mission. So, we're very focused at this moment actually in the Kubernetes project itself. We start exploring some of the open source project and in the CICD part. In addition to that, we'll start using like a few frameworks like Flux. This is one of the frameworks like GitOps in general, around like, you know, building this culture of GitOps deployment and moving toward like more ops of deployment. That's one of the areas that we're very invested in. We're exploring service mesh at this time. And I hope like, you know, we're going to get like there. Maybe next year we can talk about service mesh more. Yeah. Are there, is there something that's holding you back on service mesh? Because there's a few options out there and various maturity levels and, you know, who's driving them. But what will some of your kind of criteria be? I would say it's mainly like, you know, I'm waiting like a little bit more. I feel like 2.14 for me when we had that discussion. If we were sitting here at 2.14, we'd be discussing Mesa's via Corporate's via Swarm, right? So I think we're still at this time at service mesh as well. Okay. Any partners that you can speak to from a technology standpoint that are helping you, that you're allowed to talk about or? Well, I mean, first of all, CNCF. Yeah. I'd agree, I'd agree, I'd appreciate you know the help that's at. Most of the public, you know, cloud providers are helping us in these areas as well. Yeah. It's really interesting. I'll be interested in catching you after the show and seeing how you thought. I mean, this is, in some ways, it was a science project a few years ago and now is this robust thing. Did you bring, I'm curious, did you bring mostly engineers, mostly managers, a mix of the two? What's the- Mostly engineers. Yeah. Mostly engineers. Hands on, get- As the old hands on. I mean, this is like another change in culture right now where most of our engineers are like the next generation, who's their full stack engineers. We're using very agile process at this moment to move forward. Our road maps internally is being published. It's being used like a voting process to go like with continuous deployment and continuous feature enhancement for the teams. So it's fantastic honestly. Okay, Amar, what things does your team hope to achieve this week? Anything that is on your roadmap or on the public open source roadmap that you're waiting on? We're definitely exploring OPA at this moment. I think that's a big potential serve. So that's one of those areas. I think going through the showroom and try to see what's option we have as well, that's another area where we're going to be very interested in that. OPA, the policy agent, I mean, you talked about compliance before. Yeah. A few years ago, with a folks in the financial industry, you would have some arguments, some discussions, sometimes heated discussions about security in the cloud and et cetera, and a highly regulated industry, yet kind of maybe ironically or somewhat, maybe surprisingly for some, right very advanced in many areas, the whole industry. That's well known if you're in it. Do you still have to have discussions about compliance and security in the cloud? I guess maybe when you talk about data locality and international borders more. And that's what, we already have our own policy management like which is built in, we build it ourselves. And that's where I see the potential, like moving from building it yourself to more of like using an open source project and try to reuse it and contribute back to that open source community, like something like OBA for example. So that's like the next generation where I see it will help us as well. All right. Amar, any advice you'd give to your peers out there if they're new to the community, things you've learned along the journey so far? I would say like start small, don't boil the ocean, start with small COEs, small pilots program, look for success, look for goals. Technology is great, but don't just like move toward technology because just keep moving target towards the end, try to set like business goals for you and business like targets for your projects and that's how you're going to achieve success. Well, Amar, really appreciate you sharing Fidelity's updates. I wish you and your team the best of luck here at the show and beyond and we'll definitely hope to catch up soon. All right, for John Troyer, I'm Stu Miniman. Be sure to check out theCUBE.net for all of the coverage of this as well as all the cloud, cloud native and more shows that we have. Thank you for watching theCUBE.