 From around the globe, it's theCUBE with coverage of KubeCon and CloudNativeCon North America 2020 virtual brought to you by Red Hat, the CloudNative Computing Foundation and Ecosystem Partners. Hello and welcome back to theCUBE, virtual coverage of KubeCon, CloudNativeCon 2020. It's virtual this year, we're not face-to-face, we're normally in person, where we have great interviews, everyone's kind of jamming in the hallways, having a good time, talking tech, identifying the new projects and new wear. So we're not there, we're remote. I'm John Furrier, your host. We've got two great guests, both Kube alumni, Chris, and as the Chief Technology Officer of the CNCF. Chris, welcome back, great to see you. Thanks for coming on, appreciate it. Awesome, glad to be here. And of course, another Kube alumni who was in studio, but we haven't had him at a show, JR Stormant, Executive Director of the FinOps Foundation. And that's the purpose of this session, a interesting data point we're going to dig into, how cloud has been enabling more communities, more networks of practitioners who are still working together. And it's also a success point, Chris, on the CNCF vision, which was been playing out beautifully. So we're looking forward to digging in. JR, thanks for coming on, great to see you. Yeah, great to be here, thanks John. So first of all, I want to get the facts out there, because I think this is a really important story that people should pay attention to. The FinOps Foundation, the JR that you're running, is really an interesting success point because it's not the CNCF, okay? It's a practitioner community that builds on cloud. Your experience in the community that you had is doing specific things that are, I won't say narrow, but specific to certain FinTech things, but it's really about the success of cloud. Can you explain and lay out, for take a minute to explain, what is the FinOps Foundation and how does it relate to CNCF? Yeah, definitely. So if you think about this shift that we've had to companies deploying primarily in cloud, whether it be containers as CNCF focuses on or traditional infrastructure, the thing that typically people focus on, right, is the technology and innovation and speed to market and all those areas, but invariably companies hit this, we'd like to call the spend panic moment where they realize they're initially spending much more than they expected, but more importantly, they don't really have the processes in place or the people or the tools to do things like fully understand where their costs are going, to look at how to optimize those, to operate that in their organizations. And so the foundation, the FinOps Foundation, is really focused on the people and practitioners who are in organizations doing cloud financial management, which is being those who drive this accountability of this variable spend model that's existed. So we were partnering very closely with CNCF, and we're now actually part of the Linux Foundation as of a few months ago. And just to kind of put into context how the two kind of interact together, whereas, you know, CNCF is very focused on open source, cloud-native projects. You know, for example, Spotify just launched their backstage cloud management tool into CNCF. The Spotify folks on our end are working on the best practices around the cloud financial management of that and the standards that go along with that. So we're there to help define this sort of cultural transformation, which is a shift to now engineers happen to think about cost as they never did before, and finance people happen to partner with technology teams at the speed of cloud and executives happen to make trade-off decisions and really change the way that they operate the business with this variable pay-as-you-go engineers have all the access to spend the money in cloud model. Hey, playing check for engineers, who doesn't like that, you know? Rain that in, right? It's like shift left for security and now you've got to deal with the finance and the FinOps, it's really important. It's super important. Chris, in all seriousness, putting kidding aside, this is exactly the kind of thing you see with open source. You're seeing things like shift left, where you want to have security baked in. You know, what JR's done in a fabulous job with his community, now part of his Linux foundation scaling up. There's important things to nail down that are specific to that domain that are related to cloud. What's your thoughts on this? Cause you're seeing it play out. Yeah, no, I mean, you know, I talked to a lot of our end user members and companies that have been adopting cloud-native and I have lots of friends that run, you know, cloud infrastructure at companies. And just as JR said, you know, eventually there's been a lot of success in cloud-native and once you start using a lot of things, your bills are a little bit more higher than you expect. You actually have trouble figuring out, you know, kind of who's using what. Cause you know, let's be honest, a lot of the clouds have built amazing services, but let's say the financial management and cost management accounting tools, charge back is not really built in well. And so I've kind of noticed this issue where it's like, great, everyone's using all these services, everything's great, but costs are a little bit confusing, hard to manage. And, you know, serendipitously, you know, I ran into JR and his community out there because my community was having an eat of like, you know, there's just not good tools, standards, no practices out there. And, you know, the Finnoff Foundation was working on these kind of great things. So we serendipitously found a way to kind of work together and be under the same umbrella foundation, you know, under the Linux foundation. In my personal opinion, I see more and more standards and tools to be created in this space. You know, there's, you know, very few specifications or standards and trying to get cost, you know, data out of different clouds and tools out there. I predict a lot more work is going to be done in this space, whether it's done in the Finnoff Foundation itself, CNCF, I think will probably be collaboration amongst communities. Can I truly figure this out? So engineers have an easier understanding of, you know, if I spin up the service or experiment, how much is this actually going to potentially impact the cost of things. And for a while, you know, engineers just don't think about this. When I was at Twitter, we spun up services all the time without zero care cost. And that's happening in a lot of small companies now, which don't necessarily have as big budget. So I'm excited about the space. I think you're going to see a huge amount of focus on cloud financial management and Finnoffs in the near future. Chris, thanks for that great insight. I think you got a great perspective. You know, in some cases, it's a fast and loose environment like Twitter. You mentioned you got kind of a blank check and the rocket ship's going, but JR, this brings up two kinds of points. There's kind of like the whole code side of it, the software piece where people are building code. But also this, the human era. I mean, we were playing with cloud. I said, we have a big media cloud in Amazon and we left one of the buckets open or the switches and elemental. We're getting charged massive amounts for us cash. We're like, wait a minute, we're not even using this thing. We used it once and it left it open. It was like the water was flowing through the pipes and charging us. So, you know, there's human errors throwing the wrong switch. I mean, it was simply one configuration error. In some cases, it's more about planning and thinking about prototypes. Yeah, I mean, so take what your experience there was and multiply it by a thousand development teams in a big organization who all have access to cloud. And then, you know, and this isn't really about a set of new technologies. It's about a new set of processes and a cultural change. As Chris mentioned, you know, engineers now thinking about cost and this being a whole new efficiency metric for them to manage, right? You know, finance teams now see this world where it's like tomorrow the cost could go 3x, the next day they could go down. You've got, you know, things spinning up by the second. So there's a whole set of cross functional and it's the majority of the work that our members do is really around, how do we get these cross functional teams working together? How do we get, you know, the each team up leveled on what they need to understand with cloud because not only is it, you know, highly variable, but it's highly decentralized now. And we're seeing, you know, cloud hit these sort of material spin levels where, you know, the big cloud spenders out there are spending, you know, high nine figures in some cases, you know, in cloud and it's this material for their businesses. It's interesting. And let's just, let's be honest here is like clouds for the most part don't really have a huge incentive in offering limits and so on. It's just, you know, like, hey, the more usage, the better and hopefully getting a group of practitioners and real figures will only put pressure to build better tools and services in this area. I think actually it is happening. I think Jared could correct me if I'm wrong. I think AWS recently announced a feature where I think there's finally like quotas, you know, enabled, you know, you can do some quotas now for and billing limits at some level, which, you know, thank, it's 2020. Thank you. Yeah, thanks for that. No, no, just to push back a little bit in support of our friends at AWS Google, all those companies, you know, for a long time doing this work, we were worried that the clouds would be like, what are you doing? Are you trying to minimize commitments? And, you know, the dirty secret of this type of work and I was just talking with a bunch of practitioners today is that cloud spend never really goes down. When you do this work, you actually end up spending more because, you know, you're more comfortable with the efficiency that you're getting and your CIO is like, let's move more workloads over, but let's accelerate, let's do more in cloud, let's go out more data centers. And so the cloud providers are actually largely incentivized to say, yeah, we want people to be efficient, we want them to understand this. And so it's been a great collaboration with those companies. And as you said, you know, Google Azure started to really focus in this area and ship more features and more data for use. It's really about getting smart. I mean, you know, they're, no, you could do it. I mean, I remember the old browser days, you can switch the default search engine through 10 menus. Yeah, you can certainly find the way if you really wanted to dig in and make policy a simple abstraction layer feature, which is really no brainer thing. So I think getting smarter is the right message. I want to get into the synergy, Chris, between this trend, because I think this points to kind of what actually happened here. If you look at it, at least from my perspective, and correct me if I'm wrong, but you had JR had a community of practitioners who were sharing information, sounds like open source, they're talking and sharing, you know, hey, don't throw that switch, do this is a best practice. That's what open communities do. But now you're getting into software, you have to embed cost management into everything, just like security I mentioned earlier. So this trend, I think if you kind of connect the dots is going to happen in other areas. And this is really the synergy. Am I getting that right with CNCF? So the way I see it is, I dream of a future where developers, as they develop software, will be able to have some insight almost immediately of how much potential, you know, costs or impact they'll have, you know, on maybe a new service they're spinning up or potentially earlier in the development cycle saying, hey, maybe you're not doing this in a way that is efficient, maybe use something else. Just having that feedback loop a lot, you know, closer to dev time, then, you know, a couple weeks out, something crazy happens all of a sudden, you notice, you know, based on, you know, your FAs or financial folks searching out to you saying, hey, what's going on here? This is a little bit insane. So I think what we'll see is as, you know, practitioners and, you know, JR Spinoff's Foundation community, you know, get together, share practices, a lot of them, you know, just as we saw on CNCF kind of build their own tools, models, abstractions, and, you know, they're starting to share these things. And once you start sharing these things, you end up with a, you know, a dozen tools, eventually, you know, sharing, you know, knowledge sharing, code sharing, you know, specification sharing habits and eventually things kind of, you know, become de facto tools and standards. And I think we'll see that, you know, transition in the FinOps community over the next 12 to four months, you know, very soon in my opinion. I think that's kind of where I see things going. JR, this really kind of also puts a real, you know, spotlight and illustrates the whole developer first cliche. I mean, it's really not a cliche. It's happening developers first. When you start getting into the calculations of ROI, which is the number one C level question is, hey, what's the ROI of this project? Or I won't say cover your ass, but I mean that if someone kind of does a project and it breaks the bank or causes a, you know, financial problem, you know, someone gets pulled up to the back woodshed. So, you know, here you're balancing both ends of the spectrum, you know, risk management on one side and you've got return on investment on the other. Is that coming out from the conversation where you guys just in the early stages, I could almost imagine that this is a beautiful tailwind for you, these trends. Yeah, I mean, if you think about the work that we're doing and our practitioners are doing, it's not about saving money. It's about making money because you actually want to empower those engineers to be the innovation engines in the organization to deliver faster, to ship faster. At the same time, they now can have, you know, tangible financial ROI impacts on the business. So it's a new up-leveling skill for them. But then it's also, I think to Chris's point of, you know, people seeing this step more quickly, you know, what the model looks like when it's really great is that engineers get near real-time visibility into the impact of their changes on the business. And they can start to have conversations with the business or with their finance partners about, okay, you know, if you want me to move fast, I can move fast, but it's going to cost this. If you want me to optimize on cost, I can do that or I can optimize performance. And there's actually, you know, deeper ROI conversation that can develop now. I know a lot of people who watch the queue always share with me privately. And Chris, you got great vision on this. We've talked many times about it. We're learning a lot and the developers are on the front lines. And, you know, a lot of them don't have MBAs and, you know, they're not in the business, but they can learn quick. I mean, if you can code, you can learn business. So, you know, I want you to take a minute, JR, and share some educational knowledge to developers who are out there who have to sit in these meetings and have to say, hey, I got to justify this project, buy versus build. I mean, you can learn all that in business school when I had a CS degree and got my MBA, so I kind of blended it together. But could you share what the community's doing in saying, how does that engineer sit in the meeting and defend or justify or use some of the best practices? What's coming out of the foundation? Yeah, I mean, and we're really looking at first, what are core principles that the whole organization needs to line around and then for each persona, like engineers, what they need to know. So, I mean, first and foremost, it's about collaboration, you know, with their partners and then starting to get to that world where you're thinking about your use of cloud from a business value driver, right? Like, what is the impact of this? The critical part of that though is really decentralization where, you know, now you've got everybody basically taking ownership for their cloud usage. So, for engineers, it's, yes, we get that information in front of us quickly, but now we have a new efficiency metric and engineers don't like inefficiency, right? They want to write phishing code, they want to have phishing outcomes. At the same time, those engineers need to now, you know, have a, we call it like a common lexicon or for Hitchhiker's Guide to the Galaxy folks, a babble fish that needs to be developed between these teams. So, a lot of the conversations with engineers right now is in the foundation is, okay, what financial terms do I need to understand to have meaningful conversations about OPEX and CAPEX? And when I'm going to make a commitment to a cloud provider like a committed use discount at Google or a reserve instance or savings plan at AWS, you know, is it okay for me to make that? What, how does that impact our, you know, cost of capital? And then, once I make that, how do I ensure that I can work with those teams to get that allocated and accounted to the right areas, not just for charge back purposes, but also so that my teams can see my portion of the estate, right? And then we're having the flip side of that conversation with all the finance folks of like, you need to understand how the variable cloud, you know, model works, and you need to understand what these things mean and how they impact the business. And then all that's coming together and to the point of like how we're working with CNCF, you know, into best practices, white papers, you know, training series, et cetera, sets of KPIs and capabilities. And all these problems have been around for years. And I wouldn't say they're solved, but the knowledge is all out there. We're pulling it together. The new level that we're trying to tackle with CNCF is, okay, in the old world of cloud, you had one to one use of a resource. You're running a thing on an instance. In the new world, you're running in containers. And that, you know, cluster may have lots of pods and namespaces, things inside of it that may be doing lots of different workloads and you can no longer allocate. I have got this EC2 instance and this storage to this thing. It's now split up and very ephemeral and is a whole new layer of virtualization on top of virtualization that we didn't have to deal with before. And you got multiple cloud. I'll throw that in. They just make them out of the dimension on it. Chris, tie this together because this is nice synergy to scale up what he's built with the community, now part of the Linux Foundation. This fits nicely into your vision, you know? Perfectly. Yeah, no, 100%. Like, you know, so Linux Foundation, you know, as you're well aware is just a federation of open source foundations that groups working together to share knowledge. So it definitely fits in kind of the Linux Foundation mission of building the largest share technology investment for, you know, humankind. So definitely a good there with my kind of CNCF CTO hat, you know, on is, you know, I want to make sure that, you know, my community and, you know, the community of cloud native has access and, you know, knowledge about modern, you know, cloud financial management practices out there. If you look at some of the new and upcoming projects in CNCF, things like, you know, Backstage, which came out of Spotify, they're starting to add functionality that, you know, like originally Backstage kind of started out as this, you know, everyone builds their own service catalog to go catalog and, you know, who owns what and, you know, and all that goodness and developers use it. And eventually what happened is they started to add cost, you know, metrics to each of these services and so on. So it surfaces things a little bit closer, you know, at depth time. So my whole goal is to, you know, take some of these great, you know, practices and potential tools that were being built by this wonderful FedOps community and trying to bring it into the project, you know, front inside of CNCF. So having more projects either expose, you know, useful, you know, FedOps related metrics or, you know, be able to, you know, essentially tool themselves to quickly be able to get useful metrics that could be used by FedOps practitioners out there. That's my kind of goal. And, you know, I just love seeing two communities come together to improve the state of the world. It's just a great vision and it's needed. So, and again, it's not about saving money. Certainly it does that if you play it right, but it's about growth and people, you need better instrumentation, you need better data. You got cloud scale. Why not do something there, right? Absolutely. It's just maturity yet in a day because, you know, a lot of engineers, you know, they just love this whole like, you know, rental model just uses as many resources they want, you know, without even thinking about just basic, you know, metrics in terms of, you know, how many idle instances do I have out there? So like people just don't think about that. They just think about getting the work done, getting the job done. And if they, anything we do to kind of make them think, you know, a little bit earlier about costs and impact, efficiency, charge back, you know, I think the better the world is. And honestly, you know, I do see this, to me, it's almost like, you know, with my hippie hat on, it's like it's even greener for the more efficient we are, you know, the better the world off cloud is going to grow, but we need to be more efficient and careful about the resources that we use and consume. And certainly with the pandemic, people have virtually wanted mental health too. I mean, if people are going to be pulling their hair out, worrying about dollars and cents at scale, I mean, people are going to be freaking out and you're in meetings justifying why you did things. I mean, that's a time waster, right? I mean, you know, talk about wasting time. I have a lot of friends who, you know, run infrastructure at companies. And there's a lot of, you know, some companies have been, you know, blessed during this, you know, crazy time with usage, but there is a kind of laser focus on understanding cost and so on. And you're not believe how difficult it is sometimes, even just to get, you know, reporting out of these systems, especially if we're using, you know, multiple clouds and multiple services across them. It's not, it's non-trivial. And, you know, J.R. could speak to this, but, you know, a lot of this world runs in like terrible spreadsheets, right? And in versus kind of, you know, nice automated tools with potential APIs to use. A lot of this stuff is just done sadly in spreadsheets. Yeah, salute the flag to one standard to rally around. We see this all the time, J.R., in emerging inflection points. You know, de facto kind of things develop. Kubernetes took that track. That was great. What's your take on what he just said? I mean, this is a critical path item for people from all around. Yeah, and it's really like becoming this bigger and bigger data problem as well, because if you look at the way the clouds are billing, they're billing per seconds and down to be very fine in green detail, you know, or functions and serverless. And that's amazing for being able to have accountability, but also you get people with at the end of the month have 300 gigabyte billing files with hundreds of millions of rows and columns attached. So, you know, that's where we do see companies coming together and say, yeah, it is a spreadsheet problem, but you can now no longer open your bill in a spreadsheet because it's too big. So, you know, there's the native tools are doing a lot of work. You know, as we mentioned, you know, AWS and Azure and Google shipping a lot. There's great cloud management platforms out there. They're doing work in this area. You know, there's people trying to build their own and open source the things like Chris was talking about as well. But really at the end of the day, like this is not a technology change. This is sort of a cultural shift internally. And it's a lot like the like, you know, move from data center to cloud or like waterfall to DevOps. It's a shift in how we're managing, you know, the finances and the money of the business and bringing these groups together. So it takes time and it takes involvement. I'm also amazed. I look at the job titles of the people who are plugged into the Phenops Foundation and they range from like principal engineers to tech procurement, to, you know, product leaders, to CTOs. And these people are now coming together in the classic to get a seat at the table, right? To have these conversations and talk about not how do we reduce, you know, costs in the old IT world but how do we, you know, work together to move more quickly, to innovate, to take advantage of all these cloud-native technologies so that we can be more competitive, especially now. It's automation and I mean, all these things are at play. It's about software. I mean, software-defined operations is clearly the trend we've been covering. You guys have been riding the wave. Cloud-native actually is so important in all these modern apps and it applies to almost every aspect of the stack. So it makes total sense, great vision. Chris, props to you for that. JR, congratulations on a great community. JR, I'll give you the final word. Put a plug in for the folks watching on the Phenops Foundation. Where you're at, what are you looking to do? Are you adding people? What's your objectives? Take a minute to give the plug. Yeah, definitely. We're an open-source community which means we thrive on people contributing inputs. We've got now almost 3,000 practitioner members which is up from 1,500 just this summer. And we're looking for those who have either an interest or a need to plug into our technical advisory council to help define standards. As part of this event, the cloud-native con, we're launching a white paper on Kubernetes and how to do cloud-financial management for it, which was a collaborative effort of a few dozen of our practitioners, as well as our vendor members from VMware and Google and AppTO and a bunch of others who've come together to basically define how to do this well. And we're looking for folks to plug into that because at the end of the day, this is about everybody sort of up-leveling their skills and knowledge. And the knowledge is out there in everybody's head and we're focused on how to drive a central collection of that, be the central community for it and enable the people doing this work to get better at their jobs and contribute more to their companies. So invite you to join us. If you're a practitioner, it's free to get in there and plug in to all the bits and there's great Slack interaction channels where people are talking about Kubernetes or Venops and Kubernetes or AWS or Google or wherever you want to go. So I hope you consider joining the community and joining the conversation. Thanks for doing that, Chris. Good vision. Thanks for being part of this segment. And as always, CNCF, this is an enablement model. You throw out the soil, let the thousand flowers bloom. You don't know what's going to come out of it. New standards, new communities, new vendors, new companies. Some entrepreneur might jump in this thing and say, hey, I'm going to build a better tool. Love it. You never know, right? So thanks so much for you guys for coming on. Thanks for the insight, appreciate it. Thanks so much, John. Yeah, thank you for having us. Okay, I'm John Furrier, the host of theCUBE covering KubeCon CloudNativeCon 2020. We're virtual this year. We wish we could be there face to face, but it's KubeVirtual. Thanks for watching.