Hone Your DevOps Skills and Transform Your Organization | DevOps Experience 2023

Welcome everyone. Uh, we're here today to talk about how you can hone your DevOps skills to help transform your organization. My name is Brad Maltz.

I'm the senior director of DevOps and develop a relation ecosystem at Dell Technologies. When we're looking across our enterprises today, we're really thinking a lot about the modern applications. We still have to deal with a lot of legacy apps and that's 100% part of the business.

But the trick here is that even though we've been managing more traditional applications for the majority of our career going forward, modern applications are really becoming that new normal. And what's happening in the universe around these more modern applications is that IT operations, administrators and all other parts of it have to start to figure out how do they adapt to this world of delivering these newer modern applications while also supporting the traditional apps within the majority of organizations out there. What we're finding is that there are principles such as more of the DevOps operating model that have taken hold.

And when we look across things like DevOps, what we're finding is the individuals within the company actually have a lot of the power and control to make things successful or not. But the trick is how can you as an individual try to help yourself from a career skillsets perspective that will help enable the future of where the modern a modern application needs are. Now, with that being said today, what is everybody doing today?

So today, realistically, people are focused heavily on ensuring your traditional applications are still up and running, right? The businesses for years, for 20, 30, 40 years have been managing and driving these applications. Your daily job might actually look something like log into your servers, upgrade your servers, check the dashboards to make sure that things are healthy.

And then unfortunately, a lot of times you have to manually go and troubleshoot or manage things. You have to go and manage the underlying storage systems. You have to carve out volumes, manage snapshots, monitor performance.

You might have to go deal with your virtualization environment and deal with upgrading your entire cluster or ensuring that the virtual machines are actually not stomping on each other within that same cluster. And usually a lot of the tasks that you actually do in an IT ops world are done from a manual perspective. You're not going in and pushing a button and everything automagically works.

You're going into different element managers, different consoles, different screens to really try to piece the puzzle together to try to deliver the outcome. Whether it's something that is more of a root cause analysis, whether it is something that is more of just a deployment exercise. As we go through this transformation and as we as an industry really have to figure out how to scale up our infrastructure, we need to be able to optimize what we as individuals do on a daily basis, which should then optimize what the business around us actually does.

And in that case, IT ops really needs to transition more into an automated world. It needs to start adopting a lot of the principles around things like infrastructure as code observability, getting all the way down into the GI ops and some of these other technologies. So from an IT ops perspective, you as an individual are key within this world.

Now of course, this is where the hard part of the journey comes in. Personal growth is very hard for most people. Why?

It's because your day job is actually focused on keeping the lights on, ensuring that things can be properly deployed, triaging on a regular basis, and everything in the middle. When we actually talk to customers around things like DevOps, we start describing DevOps as more of a operating model. It's more about how the organization is pivoting to try to make sure that IT operations becomes more agile and goes down the road.

That software development has gone down for X number of years prior from a DevOps perspective. The goal here is not just to make sure that the developer is happy, that's one goal is to make sure that we as an operations organization are satisfying the needs of our development team. But if you flip that back, what that really means is that it be needs to become more of a platform.

It needs to become more standardized. And in this world, you as an individual need to figure out what does that mean for you? One of the first major tasks within that space is actually this kind of cultural movement that we always hit on.

You can't go and begin going down the DevOps kinda road if you don't have the right d n A within the organization. These don't usually start bottoms up. You can try that.

A lot of people do. If you have a few passionate people that are trying to do the right thing, that's great. The problem is executive leadership has to support you in this journey, and cultural movements have to start from the top down.

So that means leadership down through management, down through execution. Folks and operations folks need to make sure that they're adopting kind of the best practices principles from the DevOps uh, world. Now, in the cultural side of this, you as an individual, you need to kind of step out of your daily shell and you need to think differently because you are part of that machine that has to say, just because we've done it this way doesn't mean we keep doing it this way going forward.

Therefore, that cultural movement starts with you and part of your personal growth journey is to actually be able to take that step back and look at the bigger picture. The last part of this kind of picture that we're trying to go along in the DevOps journey is really the larger organizational transformation. A lot of times when people begin the DevOps journey, what happens there is they send a few people off into a room to figure out Kubernetes to learn Ansible, learn terraform, learn some of the basics.

But what happens as you scale that is the organization around you needs to fully transform. Sometimes that's a full on reorganization of the company, and that means that maybe you don't have the server admins versus the storage admins as much as you used to. You might still have server versus storage SMEs and you as an individual might be that person.

But as the full on organizational transformation takes hold, you really need to figure out what is your role in that world. You might still be an SS m e, but you might be an SS m e that actually has a much broader scope than you had before. Now in your daily life, reality gets in the way.

There's a whole bunch of stuff you have to deal with, as we've talked about. You might have to deal with the physical side of your reality. The data center, this, you know, this is all about how do you physically manage cabling and servers and storage arrays and all the things that technology has to run on.

You can't escape that. There might be other people that deal with it in the future depending on where your role scales to in that transformation. But part of what almost everybody has to deal with is the physical side of your reality.

The second problem that almost everybody runs into is actually around what I'll call the traffic jam. You have too many things on your plate. You have too many things that are blocking you from actually becoming more efficient and getting to your destination on time.

How do you actually do everything from time management all the way through backlog management? How do you make sure that you're prioritizing the right things that you have to work on? And this is really where your personal growth journey hits kind of one of the biggest snags, nine times outta 10 in your priority list.

Self-learning and training is not one of those activities that you actually focus on. Hopefully your company is actually helping you to make time for that. But to be able to get past these traffic jams, you do need to spend some time investing in yourself.

And that's what we're gonna talk about in about a slide or two. It's kind of what that investment in yourself needs to be. Now the third part of your reality is manual work.

And this is actually one of the first things I recommend for anybody is how do you take what you've been doing for 3, 5, 10, 20, 30 years within this industry and how do you actually automate that? And this journey is really about understanding what are those daily processes? What are those weekly processes that you go through where you can actually document the process and once you document it, you start automating it.

And you can use tons of tools out there to figure out how to do the automation, but you gotta start small. If you start small within the automation game, what you're able to do is you're able to find out that all those small things start to add up to bigger tasks and bigger tasks. And at some point, a large percentage of what you do actually becomes automated.

Now at the same point, you gotta be careful 'cause some tasks are not worth it to automate. If you do something once a year versus doing something on a daily, hourly or continuous basis, don't choose the thing that happens once a year. It's not worth it.

Choose the thing that happens as frequently as possible and work your way down the line from there. And just to back up, why this is one of the most important things out there from a recent study that we had done. 56% of the organizations that have the highest level of automation are watching their app portfolios more than double over the past three years.

That's a huge percentage to be able to say that you can support more applications simply because you are being, you're able to automate what's happening in the background. That is a huge return on investment for you being able to kind of invest in yourself, which in turn helps to invest in the actual company. Now speaking of investing in yourself, let's talk about where can you upskill yourself.

And there's multiple angles. The no-brainer side of the investment is actually gonna be around the technical skill side. In the technical skills game, there's multiple areas that are kind of newer and some that are actually, I would say a little bit even more traditional nowadays.

The first one that's on the bucket list that you have to kind of dip your toe in or else become even proficient at is infrastructure as code in the infrastructure as code space. You're looking at tools such as Ansible or Terraform as two examples, but there's a ton of them out there. So those aren't the only two.

The goal of these types of tools obviously is to going back to the last conversation to help take your manual tasks and automate them as quickly as possible. The trick in this space is there are simple ways to do things and there's more kind of traditionally aligned thought processes in some of these. And then other infrastructure as code tools are taking a little bit more of a a forward-looking approach to this, right?

And what you get into is a conversation of imperative versus declarative. Some tools out there, you basically are going to write the direction step-by-step. Step one is to do this.

Step two is to do this. Step three is to do this. That's a world that we call imperative.

And that's a world where basically you can very quickly spell out your existing process within the script or the module or the playbook that you're actually looking to build. The declarative world is kind of, I I, I'll call it a little bit of the newer mentality. Now, the reason this is newer is because the systems around us have not matured up until over the past, you know, X number of years to be able to support this and a declarative kind of infrastructure as code world.

Instead of giving the tooling step-by-step processes, what you're actually doing is you're telling it the end state that you expect. I expect this system to be powered on with this level encryption, with this configuration in place and all the other things to meet my needs. And that declarative call is basically converted into the system behind the scenes that understands how to go execute against all those things to satisfy your outcome that you are requesting.

The, the other part is a little bit more of a specific discussion that pops up a lot within this kind of infrastructure as code space. And this actually happens on a very frequent basis, is patch management. This is a great one that's very debatable within the world, but you can say, do I wanna keep all of my systems up to date with patches by having to continuously apply things?

Or do I just wanna say, when I expect a system to be at a certain patch level or security level, the system auto rebuilds with the proper levels two very different mentalities that fall into the infrastructure as code space. Now, when you move over to one of the next major technical areas, you obviously hit on Kubernetes. And in the Kubernetes space, everybody knows at least the word Kubernetes at this point.

If not, there are many, many experts within this space. Kubernetes is not easy and, and it's something that's very complex that even the people that are at the expert level, they're always learning new intricate things within the Kubernetes ecosystem. But the trick is to look at Kubernetes as one of the, uh, orchestration engines that you have access to, to make sure that those more modern applications are actually gonna be handled going forward.

So Kubernetes is key. And a recommendation I have here for you is just stand it up on really a small set of nodes, even on your laptop within a virtual machine or, or if you run Linux, you can do it more natively but get practice. Obviously not in the data center, but even at your local level, you have things like micro kh, K three s and many other smaller distributions that you can play with that are more localized to a small deployment.

Another area that you can upskill yourself within from a technology perspective is more in the continuous integration, continuous deployment realm. And when you come from more of an IT ops background, which is something like myself, this is where I actually grew up, is more that system admin consulting space. What you really need to look at is what does it mean to run a pipeline?

What does it mean to actually do continuous integration, continuous deployments? And in this type of a space, you have to think more like a developer. You have to think more about, well, when I'm running a pipeline, I need to run things like tests, integration tests, unit tests.

I need to understand how GIT works. I need to understand commit, I need to understand technology concepts. That might be a little more foreign.

But if you start to at least dabble in this space, you start to understand what your end users do on a daily basis. And maybe this is an area you'd be compassionate and you, you really fall more into that true developer tooling, S R E DevOps type of world at that point. A little bit less maybe in the IT ops.

But this is an area that just to be able to understand where your end users are going, you need to understand more of the C I C D approach. 'cause you probably will start using it as you progress down that DevOps journey to become more of a platform engineering team. Last and definitely not least because this technical side can go on forever, is a concept of GI ops.

And this kind of ties back into that infrastructure as code and Kubernetes space. When you start looking at the world through config files and manifests, you start saying that I wanna manage my systems, my infrastructure through my application in a way that I don't have to manually deal with lifecycle management, but I'm able to manage it through the config files. And I want a version control config files so that there is a system watching my repository, in this case GI, of all of my configuration changes and it will execute automatically based on those changes.

That's kind of the Nirvana state you want to get to when you start merging I A C Kubernetes and C I C D together, you're getting into more of that GI op space. And you're gonna look at projects such as Argo CD or something like fluxx. There's a bunch of tooling out there around this space.

But as you mature on this journey, GI ops is a great space for you to actually start to dive into. Now, technical skills are not the only space that you need to upskill yourself within. 'cause a lot of us actually are stuck in the daily world of dealing with technology, but there's more of the soft skills that you actually want to focus on.

One of the areas right off the bat is what I'll call is time management. A lot of us in technology are not good at that. How do you actually get into the world where you have time for the kind of personal growth mixed with the daily tasks mixed with more of the strategic tasks?

Time management is key in this space for you to actually kind of hone in on and grow. Another area is expectation management. We all wanna try to do the right thing or hopefully you do within your business, right?

And in that case, you need to make sure people understand what you can do, when you can deliver it, whether it's an end user leadership and or peers. And you have to learn how to manage expectations, because sometimes that can be really hard when somebody asks or requests something of you or the team, they might not understand all the requirements behind the scenes, all of the different churn that you have to go through. So you as an individual for you to be able to upskill yourself need to help set expectations better.

And this, by the way, deals with scope management because in this world, when you deal with expectations, you're always talking about what's the scope of the activity that you're dealing with. And a lot of people think of scope as something a developer deals with when they're running a backlog with a product manager. But anybody with an, it has to deal with scope of what you're dealing with, whether it's through a consulting engagement, whether it's through a basic, uh, task in the background.

You have to learn how to manage scope along with expectation management. Last and definitely not least is relationship management. Part of how you can help yourself succeed in all of this personal growth area is to actually network, connect with people, whether it's inside of your company, in the community or anywhere else, you can actually kind of meet people and learn about what other people are doing.

You need to grow those relationships because through those relationships, you're gonna find the opportunity to be able to grow your skills, your soft skills, as well as your technology skills. So I always harp on this, relationship management is key for something for you to learn how to do in your career. With all that being said, from a Dell Technologies perspective, we're always trying to help people figure out how to continue to upskill themselves, continue to grow themselves.

com to easy website to traverse on there. We have everything from podcasts to blogs to hands-on labs for free that you can spin up and go learn a bunch of technologies like we've already been talking about. com, we're gonna continue to scale that over the next months to next, uh, set of years.

And besides that, all of our podcasts and, and our, our vlogs are actually available in Dell's YouTube channel. So we have the link on the screen here that you can actually look at. And with that being said, I do want to thank you for your time today.

Hopefully you found this, uh, helpful and insightful to figure out how you can start to break away from the daily grind and figure out how to enable your company by upskilling yourself. Thank you. I.