 Live from the Mendeley Bay Convention Center in Las Vegas. It's theCUBE, covering VMworld 2016. Brought to you by VMware and its ecosystem sponsors. Now, here's your host, Stu Miniman. Welcome back to theCUBE, SiliconANGLE Media's flagship program here in Las Vegas at VMworld 2016. I'm Stu Miniman, joined by co-host John Troyer, who's also the CEO of Tech Racketing. And we're going to talk about developers and cloud native and all that stuff that maybe is a little bit different than the usual virtualization and cloudy talk here at VMworld. So welcome to the program, two first time guests. So Josh Atwell, who's a developer advocate with SolidFire NetApp. Josh, how have you not been on the program before? Just luck of the draw. All right, and Tim Carr, appreciate you joining us. Senior Solutions Architect with a head who is a channel partner here at VMware. Thanks for joining us. Well, thank you for having me. I really appreciate it. Guys, I guess start out a little bit. Can you talk a little bit about your journey? I know both of you, you've been in the VMware community for a while. How'd you get involved with kind of developer? You consider yourself, yeah, obviously you've got DevOps in your title there, but how'd you get kind of from where you were to where you are? Josh? Well, for me, starting out in a very small company and then growing through larger companies and into environments where developers were a key component of what we were delivering for. Gave me opportunities to get to know those guys, understand what they were doing. There was a period of time where I developed some applications as well. And having a bit of empathy and some sympathy in trying to create a scenario where they could develop and release more effectively, oddly enough, is now what we call DevOps. For me, I started off as a Linux system administrator, bash scripting, URL scripting, that sort of stuff, just getting those sorts of things done and what I quickly realized, I became system administrator, moved through the ranks, kind of VMware, that sort of thing, but what I quickly realized is a lot of those early skills in just scripting kind of translate to this whole, we can all start being programmers sort of situation, because scripting is programming. And I've done my best to try to make sure that everyone that I talked to is afraid of the word developer and has done just a little bit of scripting, feels that they can pick up these other tools and start to do that sort of thing in their careers. So, often this show is kind of pigeonholed as a, you know, just a sysadmin show. And when everybody, somebody starts to talk about development or any of those topics, people's eyes glaze over and they walk out of their room. So I mean, like Josh, you guys are a storage provider. So what relevance does your DevRel activities and your, the toolkits and the scripts and the things you do have to a storage admin? Well, interestingly enough, if you look at the term DevObs, that second part is ops. And it's really focused on enabling the developers and more importantly, simplifying the code deployment and actually being able to get what they develop out into the world where it can be consumed and used. That is forcing organizations to have new requirements, new expectations. Now if you look at when we ran applications before, they were on bare metal, you know, dedicated hardware. It was available and reliable. That was all that mattered. Then with virtualization and the virtualization age, it was do more with less. It was all about consolidating, right? We're still growing very fast, but we're consolidated. With the advent of cloud becoming a norm as an operating model, it changed to, I also need it faster. And now I think we're in an application evolution where not only do we need it to be available and reliable, we need it to be done more with less. We need it faster, but we need it as a service as well. And the reality is that as organizations are moving towards DevObs methodologies, as they're moving towards becoming more agile and lean and how they operate their operations, they're facing new requirements and that's putting pressure on your typical infrastructure administrators. The speed, automation, that sort of thing. Those are the drivers. Yeah, well, it's not even the drivers that the requirements and expectations have changed as a result of what AWS and Microsoft is were and the things that they're able to deliver from a service standpoint. Being able to swipe that credit card, I know we talk about that ad nauseam, but the reality is that that changes the expectation. The fact that you do not have these handoffs, you don't have excessive amounts of communication happening between the different groups. I mean, I can't set up a big giant storage array once and then I'm done for this two or three year cycle. That's not the way it's done anymore. Well, I would say that the reality is that that operating model still persists, but it is definitely transitioning. What do you get? What do you see in your question? Yeah, I think Josh hit on a really good point when he started talking about ops. The fact of the matter is the expectations of the higher level organizers of organizations really kind of get down to the fact that operations need to start understanding a little bit more about the applications and particularly the application deployment pipeline that it takes to get something from a committed state out actually into the wild. And that convergence is really what's driving this effort and also it's driving complexity into our environments in a lot of ways and a lot of new things, a lot of new things that people haven't worked with are becoming things that are expected as part of your resume. I almost would say that we're almost at an inflection point where it's to the point where people are going to need to have these skills to be marketable in the marketplace. So it's time to step outside of your box, pick up some of those skills. I mean, we're all in IT because we learned something from the start and we were good at learning, right? So we all need to remind ourselves that and not be afraid to pick up a book, not be afraid to reach out on Twitter, not be afraid to take advantage of all these great video learning tools that we have for ourselves today. It's great community. There's all kinds of options. Be a learner. That's the number one thing that I tell everybody who's interested in learning about this and growing. Yeah, I think the reality is from an administrator standpoint that those roles are having to evolve. If you look across the industry, the role of a server administrator, a person whose sole purpose is to manage a server, whether it's a rack mount server or blade server, that's not very frequently seen anymore, right? That got rolled into the virtualization administrator. Virtualization administrator was then required to handle the firmware and the maintenance and working on things like that. We're going to start seeing more of that from the network side and from the storage side as the management platforms continue to improve as integrations directly with those management platforms improve, extensibility and infrastructure continues to get better. It's going to start taking more and more of those tasks out of the picture and allowing these people who have invested time and energy into a discipline to do much higher level, higher value activities, focusing more on how do I deliver this as a service versus I'm really good at going in the UI or the CLI and configuring something. I think we're going to continue to see a depreciation and value for skills like that because it legitimately can be replaced by a small script or integration with something like Cinder and OpenStack or the native Docker volume plugin for Docker and with virtual volumes for VMware. You're alleviating the need for a lot of those skills and those capabilities, but not for someone who understands what the impacts are for those components of infrastructure. Those people are still required. Yeah, I think you had a really good point there. The experts, the true experts in fields like networking and storage, those people are always going to have jobs as far as I'm concerned. I think as we bring cloud into the scenario when we look at network connectivity, overlay networks, overlay networks still has to run on a network designed by somebody who understands how to make it fault tolerant, scalable, all of those things. So those people are always going to have jobs. I think what we're talking about is kind of the erosion of the entry level system administrator job as we see it now. You're going to have to pick up that scripting to maintain relevancy and pick up that programming, pick up that knowledge of source control and dev ops pipelines to maintain relevance. All right, so I've seen a lot of infrastructure shows that have tried to add little developer pieces of what they've done. Last year VMware had a small little separate event for kind of the developer crowd. This year I hear there's two tracks, there's developer track, there's a cloud native track kind of integrated in with it. The audience here still feels very infrastructure to me. It's very different from, say I go to Amazon re-invent, right here, first couple of years I went, I felt a little out of place, could put on a hoodie and hang out a little bit with that crowd. But how's VMware doing, do you see people that are kind of on the dev ops track? Are there many of them there that are here and how's VMware doing, can maybe start with you? Yeah, so I had a chance to go to DockerCon, which was actually a very refreshing conference to attend this year. I thought it was fantastic, kudos to them. And one of the things that I quickly realized at DockerCon was that it was people from all walks of life, developers, infrastructure people, infrastructure people who are now realizing that they're going to have to understand how to tame this wild beast that's a container and how to manage it and so on and so forth. I believe VMware is making strides in the right direction. I'd like to see more of that coming in the future. Usually as of last year, I had a chance to be a part of some of these sessions and it was all a part of the hang space last year. And I think I'd like to see just a little bit more developer driven sessions by maybe even some of the developers at VMware. I don't want to maybe know about how vSAN works. Maybe I'd like to understand, how does VMware do code control? How does VMware manage the DevOps program within their development organization? That's actually something that would get real developers excited about the platform in a way that I think that they could go on. And in my role, I actually spend a lot of time having to address that exact problem. NetApp has its own conference in-site that we'll have at the end of September at this very location. And I am leading the charge with a small group of folks on a developer cafe. Now, we recognize that a developer audience is not a strong audience at something like NetApp Insight. We don't have very many developers who are actually making applications. But what we do have are a growing number of partners, a growing number of SEs, we have customers who are doing this, and a growing number of customers who are transitioning into roles where DevOps and DevOps framework is part of their role. And as such, the efforts that we're putting in with our developer track and our developer cafe are very much focused on helping them understand how the technology is being leveraged in order to support those efforts, to support those who are creating the apps and developing those business value added assets that are being used to drive revenue to the company, how can you better enable them to provide a little bit of a viewpoint? So I talk about mechanical sympathy, being able to understand if you're a race car driver, you don't actually have to be a mechanic. But if you have some understanding about the mechanics of the engine and the mechanics of the car, you can provide more valuable feedback and provide a better opportunity for getting better performance or getting exactly what you need. So we're very much focused on creating more of a mechanical sympathy standpoint where these are the tools that developers are using, these are tools that we have integrations with our platforms, and these are things that you can use right alongside those developers so that you can work together in similar platforms to drive more success. I love that you're not really saying that ops needs to all become developers in 2026, but you're going to have to work with them, you're going to have to have these skills to take care of these apps, yeah. I think the reality is that there is not a single operations person, infrastructure, storage network that should exist without understanding some basic like source control, like leveraging Git, things like that for tracking configuration changes and all. I mean, if nothing else, like every engineer or administrator should be implementing those types of things and rolling them into a broader scope of how they're managing their environment because it's going to provide them a lot of gains and simplify later as they start to do more automation, using more tools, start getting into policy-based management. It's a very valuable skill. Josh, I love that. Was that mechanical sympathy? Mechanical sympathy. Yeah, I love that. That's my key takeaway from this whole talk. I think that when you look at DevOps as a whole, one of the things that you want to do and one of the things that the ops team can really provide for the devs are these key performance metrics and results of these key performance metrics to build a feedback loop. What you're doing there is you're building a human feedback loop, which is one of the core tenants of DevOps. Love it, key takeaway. And I attend a lot of developer and DevOps focused events now. And one thing that I've noted that is consistent is it's always about the journey of how it was so painful to develop and deploy the application. And then now we magically are deploying it multiple times a week or even multiple times a day. And the reality is that we don't talk about all the things that actually enable that. I often joke that if you look at a journey story, whether it be like The Lord of the Rings or Frodo, sure, he could have gotten there, but not without Sam. But nobody wants to talk about Sam as much as they should because he's not as interesting as Frodo's journey and what he had, right? But the reality is that the two of them had to work together in order to accomplish the goal. All right, so one does not simply walk into mortar for the people that want to take the next steps. Any takeaway advice you want to give people? Should they just start going to meetups, learn coding, what kind of advice do you give for people going forward? Actually, I'll give advice for those managing groups that I think if you are supporting developers and supporting an organization, maybe you're not going straight DevOps, but you do have these pains every time you have deployment. I recommend start embedding people cross-functionally, put your administrators and your engineers into the development teams for a couple of weeks, let them learn and understand what they're doing because the net result every single time, and this is consistent with every organization that I've seen that has continued to mature in DevOps is that they start doing that as part of their operational model and they build that mechanical sympathy and they start having a better understanding of how what they do impacts the other side. Tim, you're at the cold phase, right? You're working with customers every day. Is this a reality that you're seeing in your customer sites or is it coming soon? And kind of how would you, what's your advice for them? We're really seeing customers get to the point where we're integrating into what we call cloud teams. And I know it's a very early stage where we'll see network storage, compute engineers, all get together on one team and sometimes they're coupled with some sort of advocate for a development team, maybe a PM for a development team. That's where we're at at this point. And realistically, I think to truly operationalize the DevOps movement within big enterprise, you have to add developers and take agile methodology on and start looking at sprints for how you can bring your workflows together and how you can delivery your product. We're at the infancy stages in a lot of organizations. The most effective organizations are doing it. I mean, Josh can go on at length about all kinds of stats that came out and some wonderful reports on the state of DevOps by Microsoft and also just one by Puppet Labs. But net net, if you're going to enable your organization to deliver their platform consistently in an automated way, those organizations can recover better, they can experiment more and financially they have more market cap. And that's amazing. That's a conversation for the C-suite. The second you have a C-suite conversation about how you might want to organize your team is the second it becomes a no brainer. All right, well, Tim and Josh, really appreciate you taking the time out of your busy schedule and we welcome you to the ranks of the CUBE alumni. Thank you so much for watching. We'll be back with lots more coverage here from VMworld 2016. You're watching the CUBE. Thank you.