Delivering Managed DevOps Services – Ian Rae, CloudOps

This is texturong TV. Hey guys. Thanks for the throw.

We're here with Ian Ray. Who's the CEO for cloud Ops which is now part of abdom and we're gonna be talking about what's going on in the Cloud area with devops and the management of infrastructure and the hybrid Cloud lines and tigers and bears. Oh my in welcome to show.

Thanks, Michael. It's great to be here. So I don't think a lot of people know exactly who you guys are but Cloud Ops itself is a specialist in optimizing Cloud infrastructure.

And then Hampton has some expertise around manage devops and application development capability. So walk us through how two plus two equals something that's greater than five or more. Yeah, thanks.

So absolutely so Cloud Ops has been a specialist for for many years enabling software teams to build Cloud software. And so one of the key things we developed as a practice was platform engineering Cloud platform engineering and Cloud platform operations, and essentially the thesises that You know, if we remember Mark andresen's words of softwares eating the world. Well the way it's doing it is I would you know argue is cloud software is eating the world and as a result.

Lots of folks are building Cloud software the fundamental patterns building blocks are very very similar and that's where Cloud Ops focuses. And of course that enables frees up our customers to really Focus their development capacity on the things that are going to make them special the things that are going to differentiate them. So that's sort of where Cloud Ops is coming from.

And then Optum, you know, you know, obviously a more significant player in size and and scope and bread with the history around kojiko Pier 1. So hosting internet company and much more of a platform generalist one could say so data centers, you know. High-speed, you know networks and stuff like that.

And and so I think the intersection is really has to do with sort of the the big shift that's happening now around Enterprises who all have their own version of General generalized it needs are increasingly being driven not just to the internet but to Cloud platforms and I would argue most Enterprises are in some state of recognizing that they have to do a combination of the undifferentiated heavy lifting of it. All Enterprises need to do and looks relatively similar. And then say each of them is also sort of figuring out well to support their their specific business model and they're differentiated value chain.

How do they build the right software and out compete? And I think what we're able to do with Optimum products coming together is really address both of those. Sides of the business and those are I would argue two sides of cloud adoption that aren't often.

Let's just say clearly articulated but that we're seeing very much in the trenches sort of as part of this. I think I think it's often called digital transformation, right which is supposed to be some kind of very A one-way function of Enterprises going from some free digital state to being really a digital and Cloud native. All right.

Well that might be in the eye of the beholder, but it seems to me that Devops pretty much started from the bottom up in a lot of organizations in each individual little group kind of did their own thing and then we have a lot of different platforms out there. And now we're talking about the rise of platform engineering which to a degree is a centralization effort. Do you think as people go down that path that that makes them more inclined and kind of lean on a managed service versus kind of managing all these individual little instances of platforms themselves because right now it just seems like not many people have a managed service.

So what's gonna flip that Yeah, so I think we've seen some of the the relevant Trends, you know up close in the last few years. I mean one is a talent, right? So a tracting retaining Talent making sure the talent is working on the things that are going to be are gonna help your business win as opposed to Talent working on essentially, you know.

The underlying the same underlying problems that everybody has that that what I what I call the undifferented heavy lifting earlier, it's not that that work isn't. Hard and technically demanding, right? It's just that you know, do we really want our talented Builders to be kind of rebuilding the underlying standardized layers that most companies have to run on I think a lot of us on the on the business side of the equation would like to see the talent building.

The products and services that make us special. So I think that's really where managed Services as well as cloud services come into the equation. I would see you know, the reason why.

Cloud is so compelling is is a much faster time to Value where your teams aren't having to operate and upgrade the software. That's somebody else's problem to make sure it's delivered to you as a 24/7 utility that you can rely on and that way it really frees up your technical talent to focus on the things that are going to be Are Gonna Make You special and I'd say that you know, what we're seeing is a lot of companies essentially and I maybe update sort of Mark and recents, you know message, you know, soft creating the world. Well, I think you know cloud is cloud software is eating the world and I think most Companies are in some state of realizing that to compete they're going to have to be great at delivering in a cloud model as well as consuming in a cloud model.

Those are kind of two sides of the coin of cloud where you see Enterprises. struggling after a few years of just record adoption of cloud services realizing that you know that that in itself is a very challenging and worthy problem. But somehow they're gonna have to manage all that adoption as well as be great at delivering.

You know software defined services and I think that's the fascinating sort of Journey that's playing out in in a lot of Enterprises today. Do you think we're also reaching some sort of complexity conundrum because historically we were building monolithic applications that used to run in a Data Center and now we stick them up in the cloud and it's kind of a lifting ship now. We hear a lot more about Cloud native applications.

There's kubernetes. There's things that are in completely different in terms of the infrastructure that don't look like anything like an on-premises environment. So do you think that that's forcing a managed Services conversation?

So you're absolutely right. Is that that underlying all that very rapid adoption? Which is which has happened for good reason?

We do absolutely end up in a complexity conundrum. If we think about the way we used to run it, you know, we would build data centers and that we would use. a relatively small number of pretty standardized platforms and the vendors would have to they would have to build their products in a way that could come into your data center and run reliably with you operating it.

And and now what's happened is with the legitimate. driver of that rapid time to Value I was talking about You see a lot of the data that used to be in the data centers is now moving out into into the cloud. But the cloud isn't just one Cloud the cloud as many clouds right?

It's software as a service. It's infrastructure service. It's many platform, you know integration platforms.

And so I think the piece that not enough folks have Thought about beforehand and certainly are just starting to get you know, start start to realize how big of a challenge it is. Is that complexity? As you move from an architecture that was, you know relatively straightforward You know and you look let's face it how does downsides I mean there were always challenges around deploying, you know silos of softer into Data Centers and then sort of having them in keep the lights on mode and then eventually know these systems would degrade over time and get customized and you'd end up you'd end up in a situation where it was incredibly hard to to go anywhere from there.

So there were downsides that now we're dealing with different downsides where you know, modern Enterprise has hundreds of cloud services. Their data is is being let's just say drawn in quartered it's you know out of the data center and what's fascinating about it is that You know the checks and balances, you know, like, you know centralized it and procurement processes that used to protect the Enterprise, you know to make sure that things came into the data center in a really well structured well-managed way. Don't actually work really well in the reverse Direction.

And I would argue that a lot of the decision-making Enterprises is just not it's it's maladapted to a cloud World which is why you see a lot of Enterprises. You know having a mighty struggle to wrap their arms around, you know, the very rapid shift towards cloud. And and I think that's that's something we should have.

Not just enormous sympathy and and empathy for as a as a challenge. But I think it does come back to exactly the word. You mentioned is complexity complexity is the enemy when it comes to, you know, reliable scalable operations, and that said we have to deal with a level of complexity.

We've never had to deal with before. Do you think our understanding of hybrid cloud is kind of naive at the moment because you know when I look at it people are like well, I'm gonna have a Data Center and I'm gonna have one Cloud platform and turns out you're gonna have about 10 Cloud platforms with all kinds of different applications using different architectures. So do we kind of need to maybe reposition what hybrid Cloud really means to folks and kind of saying this is what this journey is gonna look like Yeah, so I mean it's a hybrid cloud is I think been naive and naive concept for a long time with some you know, really notable exceptions of I think some folks who have made it made it work for them.

But you know, it's I think it's been sort of an overblown perhaps, you know marketing promise that is very hard to deliver on and I would say I would extend that to both, you know hybrid Cloud extend that into sort of the promise of multi-cloud management. My sense is in this sort of ties back to the kind of the the two sides of the coin when it comes to the Enterprise, you know digital transformation Journey my sense is that Where we're going to start to see success. We're going to see correlation with Enterprises hiring more product managers and it and less project managers.

There's a there's a notion historically that you can just rely on the vendors to Define your architecture and I think get back in the data center model that worked and you'd have Enterprise architects who would be super well informed on here's the Best practices Enterprise architecture for virtualization, right? you know, this is this is again, I'd say it works then it doesn't work very well now and so what we're seeing we're seeing we're seeing Enterprises be successful when they start to think of their IT services as a very intentional sort of decisions on their part and I think that's where having a product mindset approach of really understanding. What does your business need what your users need?

and owning that sort of menu of services and productizing it to be a best fit to your to your to your Enterprise and your users. Is is something that requires a product management mindset and we haven't seen that historically as a as an attribute of it who is generally relied on the vendors to be the product managers and the vendors know what the Enterprise needs. And so that's a shift.

I mean, I don't know how much I'm addressing your your question. But that's a shift that were that we're seeing to to maybe address the so-called hybrid multi-cloud problem, which I think the vendors cannot solve unilaterally for the customers. The customers have to own it.

Hmm. Well for that matter, you know one cloud service provider doesn't even want to acknowledge that the other one exists. So how would you possibly get them the collaborate on some sort of architecture in the first place?

But my question to you is this we've been at this devops thing now for what a better part of a decade. Do you think that as we move towards more structure and then more managed services that we might be in the cusp of the democratization of devops where it becomes more accessible to a wider range of people who don't necessarily always have, you know, first class software engineering managers everywhere. Yeah, that's a great question.

I mean, I think that the to me devops also fits into a category of terms where I think we've had a very naive relationship. With over the last while you know, I was lucky enough to know some of the folks who kind of gave rise to the devops movement and I'll try to channel sort of I think the intent that is often lost in sort of the implementation in the real world. I think the intent was really driven from a recognition of rising complexity and that delivering on products and particular software product was not was no longer something you could.

apply successfully a traditional waterfall model to getting product out the door because what would happen is in the old days we do sort of our our Market. Analysis, and we come up with our product requirements document we come up with our engineering requirements and and we would we would flow that through and then sort of 18 months later. We would be, you know delivering.

product based upon those, you know starting assumptions a lot of ways back and I think what we'd find number one is that we were often wrong about those. Some of those initial inputs or they've changed over that time and I think that's really where lean and agile methodologies have been so valuable to delivering on software because it lets you address that that life cycle problem. But once you're into a lean agile delivery model You silos become your enemy and you know, we saw very clearly that you couldn't have, you know in a cloud world.

You couldn't really be successful have. Having software teams go and deliver. you know pieces of the software without understanding what the operational consequences really were we really need those feedbook back Loops to be really tight.

And so devops is less. I know there's a lot of folks who advertise, you know devops job titles, but for me, that's a bit meaningless. I mean devops is a is a way of working as a team to deliver with, you know, increasing quality and velocity some kind of product and that's where you really want to bring, you know product leadership developers.

You want to really embed your you don't want to have QA be something that you know lives halfway around the world and you sort of throw it throw software over the fence. You've got to have this all integrated and for me, it's really been about how do we build integrated teams? And why do we need the integrated teams?

It comes back to that complexity if you have a highly complex. Problem and solution then, you know splitting up all of the the teams and making and making it harder for them to communicate and collaborate is essentially going to introduce friction. And friction is what we're looking to solve in devops.

So I don't know if that addresses the question really, but I think we're we're seeing that devops is something that is really tied to delivery of product and and I'd sort of maybe loop back on the point about Enterprise it. I don't think you can do devops and Enterprise it unless you have product owners and product managers and I I think that's been something that has been a bit of a struggle is where we think we can maybe jump into devops through. you know some Enterprise framework like like safe, right and you know at the end of the day, you know, what we find is sometimes Enterprises are jumping forward and skipping some of the key steps in that maturity.

The business has to evolve alongside it as it were I'm As we think this through though, we do want to see the benefits of some level of centralization, but the developers are always like, you know, I want to be able to pick my own tools and use stuff when I want to use it. So we're basically trying to figure out how to have our cake and eat it too. So how do you strength and balance between what the developers want and what the organization needs Well, that's that is that is so key.

I think that comes back to the the where the rubber hits the road which is To me is is where where products management lives. It's that intersection of the need. And it's the the need and what and how we're delivering on that need, right?

So, how do we prioritize what our talent? Is going to work on and I think that's that's a hard problem. But to me that is the there's an art and science of product management that focuses at that.

layer, which is hard, I think for a lot of folks to see sometimes because when you have rubber hitting the road, it's hard to remember how much work has to go into the design of the road surface and and the related design of the tire and and the car that's driving those Wheels. I really believe that we're going to see a bit of a revolution around our relationship to User needs and I think it's going to be the rise of product management in the Enterprise and once Enterprises get more. Serious about product management.

I think they're procurement is going to that that Gap in terms of how procurement is designed today, and how needs are met. Is going to start to get resolved. With the introduction of product management.

All right, folks. You heard it here. I'm hoping that this revolution is televised right here on Tech strong that TV.

Hey in thanks for being on the show. Michael was a huge pleasure. Thanks so much for having me.

All right back to you guys in the studio.