Clearing the Fog Around Cloud Native

Get beyond the buzzwords and break down what cloud-native technologies are and what they promise in this Tech Barometer podcast segment with analyst Steve McDowell.

By Jason Lopez

By Jason Lopez May 3, 2021

Probably no two words are slung around the IT world more than cloud native. As more vendors and suppliers obliquely involved in cloud use the terminology, the true meaning of cloud native takes a beating. As the IT world wraps more tightly around cloud computing, one expert likes to unwind by keeping things simple and focusing on what really works.  

“When people say I want the cloud, what they really mean is I want the ability to flexibly deploy resources and reconfigure them as needed,” said Steve McDowell, senior analyst of data and storage at Moore Insights and Strategy. “That's what cloud means to people.”  

He said that the term cloud native refers to what and how that flexibility is deployed. It's an architecture, a set of practices and technologies such as containers like Dockers or Kubernetes, which enable the packaging of workloads. 

Moore kicks off the Tech Barometer Cloud Coverage podcast series, which explores the people and technologies behind cloud native.

“Cloud native is one of those terms that’s becoming more overloaded daily,” he said. “It means different things to different people. It's very contextual. Cloud native is about packaging, managing and running a workload that is sensitive to its environment.”

“If 2020 taught us anything, it's that you need to be able to dynamically reconfigure your data center based on the environment and based on what's happening in the world.”

RELATED

Cloud-Native Innovations Fuel Growth

Transcript 

Steve McDowell: Nobody really understands a partial issue, hardcore geeks, how virtualization works 15 years ago, you kinda needed to know that. And I think we're in the same space with some of these cloud native technologies, which is, you know, it seems to be everywhere, but it's still very, very early days. And the industry as a whole, I think is an early adopter, which is an unusual thing to see in our industry. But that's what's happening right now with cloud native. We're all early adopters in real time.

Jason Lopez: This is the tech barometer podcast. I'm Jason Lopez. And this is our cloud coverage episode with Steve. McDowell a senior analyst of data and storage at more insights and strategy. 

Steve McDowell: You don't know, and you don't know what you don't know unless you become an expert, which makes no sense. Right. We spent a lot of time talking about the individual technologies today, but the reality is in three to five years are going to be baked into everything. 

Jason Lopez: Right. Okay. Um, well, Steve, I wonder if you could just start us off by at least talking about what cloud native even means and I'm, and I'm sorry for the Sesame street question 

Steve McDowell: Cloud native is it's one of those terms that it's wow. You know, becoming more overloaded daily, right? It means different things to different people. It's very contextual cloud native is about packaging managing and running a workload that is sensitive to its environment. 

Jason Lopez: Okay. So that's level one, let's say, uh, let's go a level deeper. Can you expand on what you just said? 

Steve McDowell: Cloud historically has been a destination. It's a data center that lives somewhere. That's not mine, it's running on equipment that I didn't pay for, but I'm paying to use. And certainly a whole industry is built up around that. But when we talk about cloud today and cloud native starts to play a part in this conversation, cloud is more about an experience when people say I want the cloud. What they really mean is I want the ability to flexibly, deploy resources and reconfigure them as needed. That's what cloud means to people. Cloud native is a, you know, a set of things. It's a, it's an architecture, it's a set of practices. It's a set of technologies that allow me to package up a workload so that I can deploy that flexible experience. And maybe it's in the public cloud, maybe it's right here on my, on my laptop, right? In the easy example, there are things like container Dockers, Kubernetes. Now those allow me to package up the resources and dependencies I need and make a workload portable and flexible because for it, right, it's really all about the flexibility of experience. 

Jason Lopez: So in terms of the pandemic, um, how, how were businesses able to respond to changes? Was it because of the cloud? Was it cloud that enabled it to, uh, to do that? 

Steve McDowell: It is all about adapting to the needs of the organization and making sure data is where it needs to be in a way that's timely and reliable so that the business owners can consume that data and make the right sorts of decisions and have 2020 taught us anything it's that you need to be able to dynamically reconfigure your data center based on the environment and based on what's happening in the world. Most of the world's enterprises went to a remote workforce. The companies that adopted software defined, whatever, right, were the ones that were able to keep the business flowing forward. The ones that were tied to more legacy technologies are the ones that really struggled. And it doesn't take a lot of Googling through the new sources to find examples of both. And it was going to put your company out of business if you don't adopt it, probably not, but it's going to make you a laggard in your industry. 

Jason Lopez: Yeah. I've heard from people that, you know, years of data center plans were compressed into 2020. Um, you know, a lot of companies accelerated their, uh, their roadmaps. What's your sense of adoption in terms of, you know, do or die. 

Steve McDowell: And these aren't in danger of going out of business. If they're not adopting cloud native technologies, I think that's a little too scary, a little too far out, but they do lose their competitive edge and they do lose their ability to compete effectively. It's a dynamic environment and cloud native is the way now that we, uh, provide responsive IT services. 

Jason Lopez: Right. Um, I don't want to veer into a question of automation per se, but you know, the idea of it brings up, you know, how complex all this can get when operations extend beyond. And I've heard this many times, you know, what humans are capable of doing. And, um, you know, as you said earlier, and I'm sort of paraphrasing this, that, you know, expertise can take you way down the rabbit hole, but who's capable of seeing it. 

Steve McDowell: The biggest challenge from that perspective in cloud native is just the vast universe of different solutions to all of the different problems that I kind of touched on. You know, this is the world that evolved very rapidly. We weren't talking about cloud native five years ago, five years ago, we were still trying to figure out what cloud meant. It's more than just Dockers and Kubernetes, right? It's all the things around storage around manageability around integration with my non Kubernetes world. If I look at it what's broken in that world, it's that there's too many choices at the end of the day. The best approach to cloud native is a consistent management plane across both what I'm deploying from container perspective and deploying from a non container perspective. And that doesn't exist today, right? There's a number of point solutions. I Nutanix, uh, you know, conservatives that market VMware services that market HPE has got kind of their own proprietary thing going around. So there's no one answer for it. So if I look at what's needed in that space, it's more standardization, continued standardization, and that's going to come right again, we're still very early days here. If I'm an it practitioner, I have to make bets on things that will become standard. And that's still a hard choice. So, you know, for it guy, it becomes less about what are the technologies I'm deploying and what partner am I going to engage with? That's going to deliver those technologies over time. 

Jason Lopez: Well, it's interesting that idea of standardization, I talked to a technologist recently who cited things like Kubernetes, for example, opening the floodgates to having a way to manage containers, you know, in a consistent way. Um, I didn't have time to go into it then, but I wonder if you could comment on that idea. 

Steve McDowell: Sure. So when we talk about cloud native, right, it's a set of technologies that allow me to again, create a workload that can be deployed flexibly and reliably and at its heart cloud native is about taking, uh, an application and putting it in a container. And there's, uh, you know, many ways to do that. Docker is probably the most visible once I have a workload in a container that abstracts out and isolates all of the dependencies for that workload. So now I can place that anywhere. So that's a right. B is how do I place that anywhere? And then we start talking about container orchestration systems and that's where things like Kubernetes and OpenStack that allowed me to take these containers that I put my workloads in and deploy them where it makes the most sense if I makes the most sense, it could be that, you know, I have spare resources on pre-bought instances in the public cloud that are free at night. 

Steve McDowell: And, you know, I can just deploy these containers over there or, or, you know, maybe I need to migrate them from one place to the other, uh, you know, containers started in the dev ops world where I had lots of ephemeral workloads, right. I'm doing a few quick tasks and then going away. But what we've found now is that model makes a lot of sense for more enterprise application business, critical application kind of workloads is back containers about container orchestration and all of the things around that. Right? How do I take now? I have a certain subset of my applications, not all of them that are, uh, you know, hearing this architecture, how do I manage those alongside the ones that aren't there? Right. So not only do I have container orchestration, now I need to look at kind of my enterprise orchestration as well. 

Steve McDowell: So all these things have to fit together. And then, you know, around that set things like how do I do data protection, right? It's very different for containers as it might be for a more virtual machine driven workload, how do I do, uh, storage and persistent storage and, and, you know, migration across cloud boundaries. So there's all sorts of issues that need to be handled from a management plane perspective. So really, um, it's, you know, set of container technologies for isolation, it's a set of container management so that I can deploy and manage those containers. And then it's really about how do I manage data across my enterprise and in a more holistic way. 

Jason Lopez: Gotcha. Um, you know, I looked up some infographics on cloud native and they tend to look super simple. I got to say, and I looked up some other verticals and they can be mind-boggling and, and complex. And that kind of me to conclude the reason there aren't mind-boggling cloud native infographics, at least they weren't the top ones in the search that I did, uh, is because they're just too complex. 

Steve McDowell: It is, and it gets more complex every day. Um, I talked to a storage company, uh, recently acquired a container management system and they were giving me some statistics, right. In a traditional enterprise with virtual machine based workloads. I'm doing, you know, storage, operation management, operations, not transit, not storage, uh, reads and writes, right, but storage management, operations, you know, in the single digits to low double digits a day, right. And as soon as I turn on Kubernetes and containers, it goes up by two to three orders of magnitude. So instead of, you know, a dozen management operations by which, you know, it could mean attach a disc drive, detachment, destroy the amount of volume, whatever it goes into the hundreds is people move these container things, spin them up, spin them down, move them around. So all of this technology around us that is very well supported, the, the kinds of, uh, you know, it architectures that we've grown up with, they all need to change. And we're learning about that again, I think in real time, as we run applications and run into issues, or like, Oh, we better go fix that. Right. So, uh, that's where we are. 

Jason Lopez: Well, I appreciate your insights in helping us to, you know, put together, like, I don't know, an audio version of an infographic, uh, however brief, this is I'm wondering, uh, to close, how would you sum this up? 

Steve McDowell: I think, you know, again, cloud native, it's about the flexibility of experience and to really deliver that and to deliver that with value over time as an it buyer, is it administrators, nighty architect? I don't want to marry myself necessarily to a set of technologies that are tied to an OEMs hardware. I want the flexibility to run a workload anywhere. And I don't know that the OEMs are always incented to give me the ability to go run that on somebody else's cloud when I'm looking at it. What's the right kind of management plane technology for cloud native and blended environments, because we're all going to be in a blended environment forever. It's going to be the ones that are trying to solve the software defined data center problem, if you will, right. And that's going to be VMware Nutanix. I mean, those are the top dogs. Uh, you know, the other solutions are going to exist and have their sense of traction, but at the end of the day, right, I want a consistent experience across all of my resources. I'll never get that, but I'm going to go look at the folks in the industry that are driving that vision. 

Steve McDowell is senior analyst at more insights and strategy covering data and storage. This is cloud coverage, the tech barometer podcast. I'm Jason Lopez. Tech barometer is produced by the forecast and you can find us at theforecastbynutanix.com.

 

Jason Lopez is executive producer of Tech Barometer, the podcast outlet for The Forecast. He’s the founder of Connected Social Media. Previously, he was executive producer at PodTech and a reporter at NPR.

© 2021 Nutanix, Inc. All rights reserved.  For additional legal information, please go here.