Search
Follow me:
Listen on:

Day Two Cloud 106: Towards A More Open Cloud

Most public clouds are closed. That is, cloud providers make it hard for customers to move workloads and data out, and integrating services from competing clouds is fraught with complications.

But does it have to be this way? On today’s episode we discuss the notion of open cloud. The premise behind open cloud is about reducing or minimizing costs of migrating from a public cloud. In theory, open cloud lets organizations keep their options open to make changes, and should presumably increase competition because cloud providers can’t lock in customers.

Our guest is Chris Psaltis, co-founder and CEO of Mist.io, a startup building an open-source, multi-cloud management platform. This is not a sponsored episode.

We discuss:

  • A definition of open cloud
  • Why some kind of lock-in is inevitable
  • Whether lock-in is inherently bad
  • Why open cloud makes sense in a time of rapid technological change
  • A spectrum of openness among cloud services
  • What open cloud might look like for an enterprise
  • Whether open cloud can be sustained
  • More

Takeaways:

  1. There are no totally open or closed clouds. It is a spectrum.
  2. Do your homework.
  3. The game is survival of the fittest, design your organization accordingly.

Sponsor: CBT Nuggets

CBT Nuggets is IT training for IT professionals and anyone looking to build IT skills. If you want to make fully operational your networking, security, cloud, automation, or DevOps battle station visit cbtnuggets.com/cloud.

Show Links:

Mist.io

Mist.io Blog

The True Meaning of “Open Cloud” – Mist.io

Rule number one: Avoid vendor lock-in – Sean Boots

@cpsaltis – Chris Psaltis on Twitter

Chris Psaltis on LinkedIn

Transcript:

 

[00:00:01.010] – Ethan
[AD] Sponsor CBT Nuggets is IT training for IT professionals and anyone looking to build IT skills, if you want to make fully operational your networking cloud security automation or DevOps Battle Station, visit CBT nuggets dotcom cloud. That’s CBT nuggets. Dotcom cloud. [/AD] [00:00:25.600] – Ned
Welcome to Day Two Cloud, today’s conversation is about open cloud, and you might wonder what the heck is an open cloud and why do I care? Aren’t all clouds closed as far as I’m concerned? Well, they’re not. And we’ve got Chris Psaltis from mist.io to talk to us about it. Ethan, what stood out to you in the conversation?

[00:00:46.120] – Ethan
We get talking about vendor lock in. And this is a thing that I feel very strongly about, but in a conflicted way. I feel strongly in multiple directions for multiple reasons. Ned and open cloud as a topic kind of factors into this, because the whole idea is, well, wait a minute, I want to have flexibility and move workloads and do my thing wherever I want to do my thing. And these pesky clouds with their proprietary services that are all integrated are locking me in and how do I feel about that? And I’m of two minds. I’m strongly of two minds. And it’s really strange. Anyway, we have that part of the conversation along the way in this. And our guest, Chris, was just fantastic fielding all of our back and forth.

[00:01:23.710] – Ned
Absolutely. So enjoy this conversation with Chris Psaltis, CEO and co-founder of mist.io. Well, Chris Psaltis, welcome to Day Two Cloud. Before we get into the topic at hand, first, why don’t you tell the nice listeners out there a little bit about yourself and what you do.

[00:01:40.920] – Chris
Yeah, hi, thank you for hosting me. I’m Chris Psaltis. I’m the co-founder and CEO of mist.Io, where we are building an open source multi-cloud management platform. I am an engineer myself who transitioned to more business roles, let’s say so. Yeah, I have some first hand experience with the infrastructure stuff that we are going to talk about really, really soon.

[00:02:04.890] – Ned
That’s good to hear. It’s good to hear. You’re not just coming from the marketing land where everything is sunshine and rainbows and you’ve never actually touched a keyboard. So the central premise of this episode is sort of a move towards a more open cloud. And this is the term that I heard you use in a few blog posts and presentations. In fact, we were on the same not the same panel, but we were a part of the same set of presentations for Turbonomic.

[00:02:32.670] – Ned
What was it the AppOn Cloud Summit?

[00:02:35.800] – Chris
Yeah, something like that, yeah.

[00:02:37.860] – Ned
So let’s start with definitions. It could be a tricky term. What do you mean when you say open cloud?

[00:02:45.580] – Chris
OK, first of all, I haven’t invented the term, all right, so I’ve heard it from vendors, usually some sort of major public cloud vendor. And, you know, the premise of of open cloud is a really compelling one. It’s about helping you reducing or minimizing your cost for migrating off the certain cloud. So at least this is how I approach it. It’s about helping you move away when this is needed.

[00:03:21.570] – Ethan
So, OK, so if that’s your philosophy, because we are still going to have the vendor lock in discussion here today, that needs to happen. But but why? Why do I care about this? What is the big benefit for me to think about cloud from this more open perspective?

[00:03:39.520] – Chris
So the the main benefit is that you’re keeping your options open. Things change both for your organization but also in the wider technology landscape all the time. So, you know, maybe down the road you will need to change something when it comes down to your infrastructure or your application architecture. And having an open cloud or using an open cloud is really important in those cases because it will allow you to move to something else. So from from that perspective, it can be really helpful, like an exit strategy.

[00:04:19.180] – Ethan
You’re advocating using an open cloud that is go find a cloud product that is open and use that because of these benefits you get, as opposed to signing on for a cloud that would be, by this definition, not open and and then might lock you in.

[00:04:35.500] – Chris
Yeah. Yeah. Well, lock-in is part of the equation, but it’s not the the main thing. Let’s say, lock-in can be helpful at times because it practically helps you move faster, especially in the early phases of of the life cycle of the workload or the application or whatever. But, you know, lock in is good until something bad happens. And at that point, it’s probably too late to worry about it. You just have to jump right in and do something.

[00:05:13.540] – Chris
So what the way I am approaching approaching it is that, you know, if you have such concerns when you’re starting with something new, it’s good to keep in mind as like a far future scenario, what will happen if something goes wrong. And in this context, it’s not exactly about vendoe lock in, but vendor lock in is certainly a part of the equation.

[00:05:41.560] – Ned
Yeah, right. Right. When I’m building a new technology or designing a new application or infrastructure, I’m thinking about the choices I’m making at an architecture level. And some of those choices have long term impacts because they’re really hard to change once you’ve implemented them. Is there any cloud provider or cloud technology that you already have in mind that’s doing it right today? That leaves that door open for a potential migration or changes in the future?

[00:06:13.000] – Chris
Well, I don’t think that there is someone who is doing it. One hundred percent. Right. Like a classic example here is that every cloud provider is charging you for egress traffic. You know, they’re making it harder for you and putting a price tag on moving data. So there are, though, examples of services within cloud which can be more open than others. So it’s not like a binary state, it’s more like a spectrum. So some services can be more open than others.

[00:06:47.020] – Chris
And in this spectrum, you have to choose where where you are comfortable with and how can you manage this openness of whatever you are you’re using. So, you know, especially for large scale public clouds like AWS, Azure, you know, the bigger the cloud, it’s probably the more difficult to be more open on average at least because they have like one million services. It’s like it’s impossible for every service to be really, really open.

[00:07:22.760] – Chris
There are there are some services which are inherently more sticky, like, you know, IAM, for example, or something like that. For smaller players with a smaller amount of service offerings, it might be easier like for Linode, Digital Ocean, things like that. But in any case, you have to look at it on a service by service basis and not like cloud as a whole. It’s really hard to to talk about the entire cloud being open.

[00:07:52.290] – Ned
And it sort of sounds to me like if you’re looking at some of the more we’ll call them advanced or platform as a service offerings from the major cloud vendors, those do tend to lock you in a little bit more. If you’ve developed everything for AWS Lambda or you’ve developed everything for Azure Cosmos DB or something, it’s going to be harder to move off that. But if you just stay on their IaaS offering and it’s just basic virtual machines, that is easier to move off on. But you do miss out on some benefits. So how do you how do you decide on that trade off?

[00:08:25.960] – Chris
It’s really hard. It’s really hard. You know, there is there are no really good answers here. And I’ve seen it going really wrong for many people on both sides. I mean, both like with. Lambda example as well as OK, let’s put it let’s put everything on VMs, right?

[00:08:44.320] – Ned
Right.

[00:08:45.010] – Chris
So I’ve seen organizations, for example, trying so hard to avoid vendor lock in that they’re putting VMs on top of those VMS are putting kubernetes. And like all of that for the future possibility of running kubernetes on another cloud provider or on Prem or like installing some weird exotic application on the control plane level and and without having any prior experience with Kubernetes.

[00:09:13.870] – Chris
Right. So they’re trying to do like from zero to self hosted Kubernetes on the cloud. But like in this case, it’s not exactly a cloud. It’s just like somebody else’s VM. So it can go really wrong in both directions. What what you need to do is check what are your priorities as an organization, what are the available resources that you have, the skill sets and then adapt accordingly? I mean, it would be much simpler for the example that I gave you before, to start with, like some some sort of managed service where you don’t have to worry about the control plane, start breaking down your applications in micro services, figure out how to build the VMS, how to build images, how to deploy all that.

[00:10:01.420] – Chris
And then later down the road and when the team and the workload is more mature, then worry about how I could move the control plane somewhere else.

[00:10:12.640] – Ned
I think you made an important distinction for me right there that that that jumped out, which is if you’re going with a managed service like Kubernetes as a service or something, it’s not necessarily like proprietary to the cloud. The thing that they’re doing is the management layer, which is the part that is not super important to you early on. And it’s not something you’re going to there’s not a whole lot of difference in managing it yourself versus letting the management. But you still maintain control over the kind of the I guess, the customer control plane and the data plane of things. Is that kind of how you’re thinking of it?

[00:10:48.130] – Chris
Yeah, yeah, this is it. But, you know, back to the initial discussion about a vendor lock in right now in this situation, you you’re already locked into things like the managed service, EKS, AKS, GKE, whatever, and kubernetes itself. And even if you take out the managed service and you host it somewhere on Prem, let’s say you’re still locked in Kubernetes. Right. So there’s still lock in there in some in some way.

[00:11:18.460] – Ethan
We’re saying locked in. Right. But but are we locked in or are we just committed to a certain path for a given amount of time? So many of the technology migrations that I’ve done, we were locked in for some amount of time. And so the way to get out of that lock in scenario was to build the new thing, whatever the new thing was, then migrate over to it. And I’m an infrastructure guy. So most of the things I’ve been those are infrastructure, right.

[00:11:42.790] – Ethan
We’re moving to new core switches, we’re moving to a new storage array, whatever it is. And then so you you come to this, I don’t know, Flag Day, whatever you want to call it, the big change window. And we’re, we’re cutting over, it sucks. It was inconvenient. There was maybe training involved, but we weren’t locked in. It was just horrible for a few months while we transitioned over. Sometimes I feel like we used lock lock in too much, you know, like we’re here forever.

[00:12:11.890] – Ethan
And I think we can be locked in depending on what the product is. But for a lot of things, it’s more like it would just be super inconvenient to move. But we’re not locked in.

[00:12:20.380] – Chris
Yeah. Yeah, exactly. Exactly. I think like, yeah, you’re you’re right. Maybe the term is not very correctly used to be honest. Like I, I approach the, the lock in term as I suspect from again like us in the open cloud situation there, there is no real lock in. Never there’s nothing that’s one hundred percent locked in. You’re always have options. Right. It’s just a matter of how expensive are those options and how expensive it is and how big of an effort or inconvenience it will be needed to do something else.

[00:12:52.570] – Chris
So it can be like a very light version of lock in, but it can also be a very expensive one. I don’t know if lock in or commitment or something else is a better term to describe that, but I’m I’m mostly approaching it from the perspective of what will it take to take this thing out, throw it away and move somewhere else. And by the way, this is why there are no there no. Not no. But like there are very few successful migration projects going on.

[00:13:26.830] – Chris
That’s why there are still mainframes running somewhere inside your bank.

[00:13:30.860] – Ethan
Yes. Yes.

[00:13:32.080] – Chris
And, you know, like at some point you just have to do the math. Like, would it make sense to migrate off this platform or not? And how.

[00:13:40.910] – Ethan
And sometimes the answer is no, exactly. There was some government agency running, I think a Social Security Administration application on some ancient Unisys minicomputer. Did we migrate it? The cost to migrate it was astronomical and the program was being phased out anyway.

[00:13:57.860] – Chris
So we just left this thing sitting in the corner doing what it did and scrounge some spare parts just in case the mini computer died for some reason to keep it going, because that was – in that situation – the smartest thing to do. Were we locked in? You bet. But there was an end to it in sight. So what was the point of migrating over? But I am still thinking about the whole lock in concept. I’m I’m of two minds about it.

[00:14:19.760] – Ethan
I’m you know, in the idealist world, I don’t want to be locked in. I want to be able to mix and match. Buy what I want, use what I want, get the best pricing and, you know, make those decisions. Practically speaking, that doesn’t actually work. And Ned, I think you mentioned skill set this you got people that know what they know and moving to something else, even if conceptually you can get to that other vendor that offers the same service, the way that other vendor delivers that service could be so different.

[00:14:49.040] – Ethan
It can be a fairly significant transition to get your team skilled up so that you can use that. That’s not lock in. But again, goes back to that inconvenient thing. Then there’s the issue of vendor synergies where you use more than one of their products. So they all fit together real nice. And as long as the cage that got you locked in is decorated well, this ain’t such a bad place to be. I don’t think I mind so much.

[00:15:12.600] – Ned
So I got to take issue with that because I had used vendor solutions before that are supposed to all work seamlessly together and they don’t. It’s usually. So at least in my personal experience, I’m not going to throw any vendors under the bus. I’ll be nice. But in my experience, what what happens a lot of time is it’s, A it’s different product groups developing these products and B, a lot of the time it’s products that they acquired from another company.

[00:15:41.390] – Ethan
That.

[00:15:41.540] – Ned
And so the integration that is touted by the marketing material is really not there. When you actually go to implement it. It’s you can see all the duct tape and all the PowerShell strips that are running behind. And…

[00:15:56.750] – Ethan
I think they change the header graphic and that’s it.

[00:16:01.490] – Ned
Seriously, look at the executable name. Is it still the same executable name from like three products ago? OK, so I think that that can be a red herring. The idea that all the vendor products are going to work well together. They, they can. I’m not saying they can’t, but a lot of the time that that is not necessarily the case. So it makes more sense to pick the thing that works best for you from a business perspective. But I guess the thing that I always get stuck on is when do you decide to make a change?

[00:16:31.220] – Ned
What is the inciting event that makes you go, OK, this is no longer working for me. I need to shift to a different technology or a different cloud or a different provider. Chris, what do you think is some of the reasons you would decide to make the switch?

[00:16:45.950] – Chris
Yeah, one obvious reason is having your CFO knocking on the door and saying, like, you know, it’s time, it’s time to move away from that. It’s very expensive. Right. That’s not not a good reason to move. But I think that you you you understand when the time is right, as long as you’re looking actively to detect any any messages coming your way. So there’s no like recipe. It’s about always being aware of the environment changing, the business needs changing, the team changing and being able to capitalize that and do the best that you can with whatever you have at hand or whatever your business needs are at that point.

[00:17:40.970] – Chris
So it’s really it’s really hard to get the timing right. You just have to be alert. I mean, that’s that’s probably the only the only way to do it.

[00:17:49.750] – Ethan
Well, you mentioned business needs changing and so on, but then also technology needs changing. Could be the vendor you’re in just kind of fallen behind a bit. They’re doing rent extraction. They’re not really updating the product much. And you see the new thing and someone that’s in the competitive space is like. Oh, if we moved over there. We could, you know, save money or be more efficient or whatever the technological advantages. And you really have an incentive to migrate.

[00:18:13.940] – Chris
That’s certainly a factor. And especially if you consider how fast the things are changing right now. So, you know, as an example that I use very often is we went from VMs to docker and containers to Kubernetes to Serverless and like one hundred different shades of that in between. In like what, like seven years? When was the original announcement of docker? Like seven, eight years, something like that. So. And this seems like a lifetime, but it was just that eight years.

[00:18:51.150] – Chris
It’s it’s probably within the scope of your medium term planning in a big organization. And contrast that to how long it took to go from the mainframes that we were discussing before to VMs. So the the velocity of innovation in the infrastructure space is is staggering right now. And I don’t know what’s coming next. And if you are not in a position to follow the next wave, the next big thing, adopt and evolve, adapt and evolve, then you will probably lose the game at some point.

[00:19:27.690] – Chris
So it’s not so much about the technology itself, but it’s more about the culture within your organization, having people in your team who are willing to learn, experiment, adapt, move forward, try new things out and also be business aware, like what are the business needs? Like do, I do that just because it’s cool or is it actually solving a problem for my business? So, yeah, that’s that’s probably the only way I can think of where you can be.

[00:20:01.340] – Chris
Absolutely. Future proof, let’s say having like a strong team, having a strong team that can learn, adapt and move quickly.

[00:20:12.820] – Ethan
So, Chris, this is the Day Two Cloud show, and very often when we’re talking cloud, it’s contextually what we mean are probably the big three public clouds. AWS, Azure, GCP.

[00:20:23.140] – Chris
Yeah.

[00:20:24.220] – Ethan
As we’re talking through this vendor lock in stuff, one of the things that pops to mind is all of those big three public clouds have lots of other add on services that again, in theory – Ned – in theory they all work together really well. And, you know, there’s a synergy to be gained by using them.

[00:20:40.070] – Ethan
So are you talking, Chris, or would you suggest that if I’m a big three public cloud consumer, that I maybe don’t leverage some of those services that would lock me in if I can avoid it and try to keep things simple? So I’ve got more of that flexibility?

[00:20:55.900] – Chris
Yeah, it depends how hard that could be. I mean, you cannot avoid IAM, for example. And that’s the biggest sticking point on all the cloud services. Right. So it depends. It’s really hard to tell. In some cases, it could be fairly simple. I mean, instead of CloudFormation, use Terraform or something like that or instead of like another weird, very cloud specific technology that has a really good open source alternative, I would just go with the open source one.

[00:21:26.410] – Chris
I mean, the APIs are still there. There’s no there’s no reason to to pick whatever a certain cloud provider offers. But in other cases, it’s just unavoidable. So it totally depends on the case.

[00:21:40.870] – Ned
OK, so would you advocate for using an open source technology, even if it’s a managed offering versus something proprietary that the cloud is built? For instance, I’m thinking of using KNative as opposed to using Lambda for your serverless or what’s the database technology that that AWS kind of, I won’t say stole, but they they launched relatively recently.

[00:22:08.110] – Chris
They forked you mean?

[00:22:09.270] – Ned
Yes. Yes.

[00:22:12.070] – Chris
So I think that you have to do take one step at a time. That’s probably the safest approach. Don’t go all in with the latest, coolest open source project that you’ve seen out there, like take it one step at a time and I will go back to this managed Kubernetes example, which I think is a really good one. Kubernetes is a very strong platform, is very popular. I don’t know if it’s needed like in every case that it’s being used for, but that’s another story, probably for another show.

[00:22:47.410] – Chris
So, yeah, but in any case, you know, starting with your own Kubernetes cluster. It’s it’s a hard it’s a hard thing to do. So what’s what’s the easier path to running Kubernetes, starting with a managed service? Right. So there you have this combination of proprietary, let’s say, platform. But a platform that’s offering you value right out of the door and value over like a long period of time, maybe forever, instead of trying to build everything from scratch yourself.

[00:23:22.090] – Chris
And then, you know, who knows, maybe two years from now, you decide to ditch Kubernetes all together or you decide to move Kubernetes on Prem to for whatever reason. So, you know, it’s a you can take it one step at a time. You don’t have to go all in with something. To the KNative example. That’s a much tougher one because, you know, you need to have a lot of things going on in the background. You run your own KNative stuff. So I don’t know. This is this is a harder.

[00:23:53.500] – Ned
Right. Right. I mean, to a certain degree, if you’re using Lambda, you’re probably writing the application in Python or JavaScript or whatever language you choose. You can port those functions over to a different Serverless platform.

[00:24:07.000] – Chris
Yeah, I don’t know. In theory. In theory it works like. But then you maybe this is probably the easiest part that you just described, rewriting everything or porting the code itself. But then you have the whole operational level, which is a totally different game. It’s how do you deploy stuff? How do you monitor all that? How do you do logging like it can? It can be crazy. It’s it’s another it’s a totally different world.

[00:24:39.790] – Ethan
[AD] We paused the episode for a bit of training talk training with CBT nuggets. If you’re a Day Two Cloud listener, you are you’re listening to it right now, then you’re probably the sort of person who likes to keep up your skills, as am I. Now, here’s the thing about Cloud is I’ve dug into it over the last few years. It’s the same as on Prem, but different. The networking is the same, but different due to all these operational constraints you don’t expect.

[00:25:03.850] – Ethan
And just when you have your favorite way to set up your cloud environment, the cloud provider changes things or offers a new service that makes you rethink what you’ve already built. So how do you keep up with this? Training. And this is an ad for training company. So what do you think was going to say? Obviously training and not just because sponsor CBT nuggets wants your business, but also because training is how I’ve kept up with emerging technology over the decades.

[00:25:25.360] – Ethan
I believe in the power of smart instructors telling me all about the new tech, so that I can walk into a conference room as a consultant or project lead and confidently position a technology to business stakeholders and financial decision makers. So you want to be smarter about cloud CBT Nuggets has a lot of offerings for you, from absolute beginner material to courses covering AWS, Azure and Google cloud skills. Let’s say you want to go narrow on a specific topic. OK, well, there’s a two hour course on Azure security.

[00:25:55.240] – Ethan
Maybe you want to go big wide. All righty. There’s a forty two hour AWS certified SysOps administrator course and lots more cloud training offerings in the CBT Nuggets catalog. I gave you just a couple of examples to whet your appetite. In fact, CBT nuggets is adding forty hours of new content every week and they help you master your studies with available virtual labs and accountability coaching. Interested? Of course you are so satisfy your curious mind by visiting CBT nuggets, dotcom slash cloud and figure out if CBT nuggets will work for your training with their seven days free trial.

[00:26:31.540] – Ethan
Just go do it. CBT nuggets dotcom cloud for seven days free that CBT nuggets dotcom cloud. And now back to the podcast I so rudely interrupted. [/AD] [00:26:44.160] – Ned
The next big battleground that that we’ve seen looming on the horizon is the edge and Edge Cloud we recently had.

[00:26:52.650] – Chris
Edge is Coming.

[00:26:53.190] – Ned
I know we had a really good chat with Alex Marcham about Edge cloud. And I’m curious to hear your thoughts. Is there an open cloud version of edge computing?

[00:27:04.090] – Chris
Is there an open cloud version of anything that’s being something something so unique? I’ve listened to the episode, by the way. It was it was a really good one.

[00:27:15.240] – Ned
Well thank you.

[00:27:17.880] – Chris
Yeah. There’s no there’s no there’s no such thing. But this is not specific to edge like the edge certainly has some different characteristics, let’s say, when compared to the public cloud, which probably make the situation a little bit harder because maybe you also have to account for hardware and or rather special hardware or maybe you’re just using a hyperscaler extension to the edge and edge comes in all sorts of ways.

[00:27:52.810] – Chris
So it’s really, really hard to.

[00:27:56.940] – Ethan
Can be different depending on the use case and depending on who the customers are that are going to consume that edge cloud, yeah.

[00:28:02.310] – Chris
Yeah. It’s a super hard like I don’t know that the I don’t know if there will ever be a very open solution about it.

[00:28:11.490] – Ned
Yeah. That’s there’s a lot of confusion over what edge even is at this point. It’s so hard.

[00:28:16.320] – Chris
I don’t I don’t think I know still I’m struggling with edge probably more than cloud.

[00:28:22.650] – Ned
Yeah, I don’t know. Let’s let’s let’s let’s apply this to a potential like real world example. Right. Let’s say that I’m a CIO at like a fairly large enterprise and we’ve got workloads running in Azure and AWS and something on my on prem data center. And you’re advising me, Chris, you’re my advisor. You’ve come in and you’re ready to tell me what what should I pursue? I’m looking maybe two years into the future. I think any further than that might might be foolish.

[00:28:51.930] – Ned
But let’s say I’m looking two years into the future. I want to make sure that I’m prepared for that future. What would you advise me to do from a cloud perspective?

[00:29:03.670] – Chris
Hire the best people that you can and let them worry about it. That’s probably the only good solution that I can think of. But, you know, like jokes aside, like other than that, you really have to dig deeper into what are you currently doing and what is a top priority for your business. It’s not about technology. It’s more about the business. What does the business need? Does it improve? Does it need to improve its bottom line?

[00:29:34.910] – Chris
Does it have cash to spare and invest in capital expenditures? Where are the customers? What are the applications? Are there any legal restrictions or constraints or what’s the situation like? And now the hard part is projecting that in the future.

[00:29:53.870] – Ethan
What is this cloud and capital expenditure of which you speak?

[00:30:00.530] – Chris
You know, it’s probably you’re not as as hipster as I am, but I’ve seen a lot of people talking about moving workloads on Prem again. So, you know, maybe some people will pursue that.

[00:30:18.240] – Ethan
Well, OK, so so Ned’s question here, this scenario was planning a two year time horizon. And Chris, the answer you gave in this consultancy role was basically says things are changing so much. You know, you jokingly said hire good people. But I think there’s a lot of truth in that. And the idea being things are changing so rapidly to best position yourself, you need to be very closely in touch with the market where things are at, what the product offerings are and how they align with your business so that you can even, I mean two two years is not a long time horizon.

[00:30:54.470] – Ethan
Historically for IT purchases. A lot of times you’d be you would be investing in on Prem and some of that gear would have at least a three, if not five, seven or even 10 year lifespan. So to not be sure what you’re doing over a two year lifespan is really speaks to how quickly the market is changing.

[00:31:11.960] – Chris
That’s the main problem. Exactly. Because of this rate of change, it’s really hard to be absolutely sure for specific technology or something like like that, projecting projecting it in like two, five, 10 years from now, like 10 years is probably out of the question altogether. But and that’s why you you need to build this internal, let’s say, capacity and culture of change. That’s not technology related so much. Obviously, like there are some technologies that are more favorable in cultivating this culture, like open source, contributing, taking part in the community and things like that.

[00:31:58.100] – Chris
But, you know, it all comes down to the people. The technology doesn’t run itself, doesn’t run. It doesn’t run itself yet. Right. But the people do. So you need to be absolutely sure about the team that you have.

[00:32:11.660] – Ethan
Well, let me flip Ned’s question kind of on its head and ask it from a different corporate perspective, Ned. You are asking from the more established company CIOs, big plans, thinking for the future. OK, I’m a I’m a tiny startup with maybe 10 people building out software service and stuff. Should I be thinking about open cloud and being able to have that flexibility and move things around? Or should I just like go all in on AWS like everybody seems to and just hope for the best?

[00:32:40.820] – Chris
Look, I’ve been there. So in this situation, you’re probably worrying most about bringing food on the table rather than vendor lock in, right? Yeah, the situation is certainly different there. It’s not entirely different, but it’s a lot, it’s a lot different. And, you know, in the early phases, you you’re trying to prove that whatever idea you have in mind works. So you need to move fast. You need to reiterate. You need to improve stuff.

[00:33:11.450] – Chris
You need to change. But then what happens if, like one and a half year down the road, you’re extremely successful and your a serverless bill has raised like 10K per month and you can not really re architect anything right now because at that point. This is your money making machine, you cannot stop it. Right? I’m not I’m not saying that you should carefully plan about every exit scenario possible out there, but it should be at least like somewhere in your mind that what do we do when X scenario happens? Right. It’s certainly not a low priority yet.

[00:33:55.630] – Ned
You can get into analysis paralysis if you’re trying to think of every potential outcome. But I think you’re right. If you’re thinking. Hopefully you’re planning for success and maybe even wild success. So what’s going to happen if if you are wildly successful and suddenly your application does scale out to, you know, tens or hundreds of thousands of nodes, what does that bill look like? And are you going to be able to pay it with what you’re charging today?

[00:34:23.350] – Chris
I don’t know. Yeah, that’s that’s a that’s a big question. And that’s like within the frame of planning two years ahead. Right. Right. I don’t know. I wouldn’t I wouldn’t worry about it and I didn’t worry about it. But those were different times, to be honest. So you didn’t have so many options and having so many options is good, but it can also be really hard sometimes.

[00:34:48.770] – Ned
Right, right. Right. It’s funny because if you think like maybe two years back, Zoom was not a very big company. Then something happened. Suddenly Zoom had to scale up a little bit, and that’s certainly true of things beyond Zoom. I mean, like I see all the kids with their Tik Toks and TikTok was not a giant thing, but they’ve had to scale up tremendously as well. So you can be a runaway hit and that can cause you a whole new series of headaches.

[00:35:17.010] – Chris
Yeah. Or it doesn’t have to be just like this huge success, because obviously, if you’re TikTok and Zoom, you don’t care about anything else rather than the success itself, this can solve everything. It’s like a very good problem to have. But it might just be that whatever customers you have have different needs over time when it comes to availability or scaling or whatever. So it it will change. Your audience will change over time and your application and your infrastructure should go hand-in-hand with that change. That’s that’s pretty much a more realistic scenario let’s say.

[00:35:57.540] – Ethan
I feel like we’re giving people an impossible task, given infinite variables and an unpredictable future, both of your company and the technology. Make sure you make the right choice now. Got it. OK good. Well, let me ask a more practical question Chris. If I’m if I’m do I want to do open cloud at that? That feels like the right solution for me. I want to do that and maintain that flexibility in my infrastructure. What does that do as far as managing everything?

[00:36:23.070] – Ethan
Because one of the benefits of cloud, the context here is less undifferentiated, heavy lifting, doing management things I don’t want to do. I can kind of push that on to cloud the deeper I go. If I go open cloud, does that shift some of that managerial burden back to me?

[00:36:41.460] – Chris
Not necessarily, and that’s the good thing about the open cloud, let’s say, because there are several open source technologies that are offered as a managed service right now. So it this takes away a lot of the management burden. Again, back to the K8s example or containers as a service or the KNative thing or whatever. So there there are a lot of options where you can go with an open source project and you kind of go on rather open cloud service without increasing your management burden.

[00:37:20.490] – Chris
In fact, it’s the other way around. If you try to sell folks, it will be it will be a terrible pain.

[00:37:26.010] – Ned
Right. It seems like at a certain scale sometimes bringing it in house and managing yourself does make sense because there are some benefits you could get when you’re running at scale.

[00:37:36.220] – Chris
Are you referring to Dropbox?

[00:37:37.740] – Ned
I might be referring…

[00:37:38.880] – Ned
Well, I mean, that is the canonical example, isn’t it? I almost feel like it’s a black swan sometimes, so I don’t think we should go too far down that rabbit hole. But yeah, there definitely is something to making that decision. And maybe what the bigger point is, you have to reassess on a regular basis. You can’t just make a plan. This is the plan. This is what we’re going to do for the next six months or the next year. You have to have a reevaluation every few months to are we still on the right path? And maybe you are. Nothing has to change, but maybe you aren’t. And you have time to make adjustments before things get completely out of hand.

[00:38:21.300] – Chris
Yeah, sure. And as a workload or application matures and the relevant infrastructure and your team then this opens up more options. The more experience your people get and the more experience you have with running a specific workload, then this might open up possibilities that are already existing out there. And you didn’t consider, consider them when you were starting. So, yeah, it’s it’s a constant process. It never ends. And that’s the beauty of it.

[00:38:55.520] – Chris
Otherwise everybody would be bored.

[00:38:58.800] – Ned
But we seem to keep keep coming back to the people you need to have the right people in place. So when an organization is trying to support this type of initiative, what type of person are they trying to hire? Are you looking for a cloud architect or are you looking for a developer who knows a little about it infrastructure? Are you looking for someone who’s really deep into storage or networking or something like that? Or is it all of the above? Like what type of person am I trying to hire for for these roles?

[00:39:29.080] – Chris
It’s hard to to give you an answer here based on the roles that you mentioned, because it greatly varies depending on what you’re trying to do. Obviously, having a strong skillset is is a plus. But I believe that the key to success is softer kind of skills like capacity to learn, curiosity, or communication skills. These people will not work in a vacuum. They will work as part of a team. And having such qualities in your team is probably more valuable than having great technical skills and just that.

[00:40:14.380] – Chris
Right. So I would I would probably prioritize those softer skills, to be honest.

[00:40:22.320] – Ned
Interesting.

[00:40:23.640] – Chris
Yeah. Over like pure technical know how.

[00:40:28.480] – Ethan
You say interesting Ned, are you disagreeing because I actually I agree with Chris here.

[00:40:32.970] – Ned
I like the point. It’s almost like we said, OK, the technology isn’t the important thing, it’s the people. And it’s not even the technical skills of the people. It’s the people. If you have people with solid soft skills, I mean, they obviously have to be, you know, intelligent and good at their job.

[00:40:49.890] – Ned
But it’s really about those soft skills and being able to deal with each other, being able to learn new things and being able to interact with non-technical folks, that all that seems to take priority over what technical stack you happen to be running in a public cloud.

[00:41:06.990] – Chris
Yeah, because in this fast changing environment that we describe, it ends up being a game of survival of the fittest. So what do you want to do as an organization in this in this game is being able to learn, adapt, evolve, otherwise you will you will die. Right. So and this is only dependent on people. It’s not so much dependent on technology.

[00:41:32.940] – Ned
Yeah. I think this is like an important thing for listeners to to take for themselves is if you’re going to develop anything. Develop those soft skills first, I mean, obviously understand your job and what you’re doing, but if you’re looking for a way to advance or move up in an organization or just improve what you do, work on those soft skills and and try to develop those, especially communication and working with others, as opposed to just grinding out the technology in another certification.

[00:42:04.200] – Chris
And these ones are that hard to find and the hard ones to develop. Right. OK, I mean, cloud’s not rocket science. Everybody was like reasonably technical. At some point we’ll understand it. Maybe not the pricing model, but everything else, OK?

[00:42:22.620] – Ned
Right. No one understands the pricing model, not even the cloud providers. Well, I guess the last thing I wanted to to bring up is how sustainable is this effort that we’re talking about to build the open cloud? Am I going to be working with all these community supported open source projects or is there going to be someone I can actually pay to to give me support? Because I know at least from an enterprise perspective, there’s an expectation that you can pay someone and when you have a problem, you can call them.

[00:42:57.540] – Ned
So is that something that is included in the open cloud that we’ve been talking about?

[00:43:03.480] – Chris
Well, if you’re open cloud services powered by an open source project, then you will probably have to do both. And, you know, all reasonably popular open source projects have some sort of enterprise behind them, supporting them in a professional way. So it’s not such a big deal. I wouldn’t worry about that a lot. And, you know, part of this whole community engagement will also be beneficial for your own teams, your own team members, because they will they will be able to interact with maintainers, maybe contribute some code, get to learn the platform they’re using better even influence its development, although this is a little bit harder.

[00:43:46.980] – Chris
But, yeah, it can happen. But in any case, it’s certainly sustainable. We’re not talking about like an obscure open source project by someone writing the entire code base alone. Somewhere in the in the world we’re talking about more popular and more. Popular and well adopted projects.

[00:44:11.680] – Ned
Right, I mean, aside from open SSL, which is like one guy.

[00:44:16.780] – Chris
Oh my God yeah.

[00:44:17.580] – Ned
The whole internet uses it. It’s more than one guy, but it really is like two people who maintain that it’s a big mess.

[00:44:25.690] – Chris
This is a little depressing. Yeah.

[00:44:27.850] – Ned
I think an important thing to draw out on that is, like you said, you can influence how a product is developed if you’re more involved in the community and make it tailored a little bit more for your particular use case, because I know that especially for an open source project, they’re always listening to the issues people bring up. And even if you’re not a developer yourself, you can log in issue or make a feature request. And if enough people do it, then that ends up in the product, which is a virtuous cycle as far as I can tell.

[00:44:57.390] – Chris
Yeah, yeah. That’s a win win situation for everybody involved, both the open source project, the community, as well as your team and your organization. So, yeah, it’s it’s something that’s really, really exciting and gives you a new perspective on things. It’s I would strongly recommend participating in such communities whenever whenever possible.

[00:45:20.980] – Ned
Not only that, but it keeps your ear to the ground about new technologies that might be coming down the pike that aren’t going to be a fit in three months, but they might be a fit in that magical two year, two year horizon we keep talking about.

[00:45:32.380] – Chris
Exactly. Exactly.

[00:45:34.600] – Ned
Awesome. Awesome. Well, Chris as we’re starting to wrap up, what we like to do sometimes is have the guest give us a few key takeaways. So do you have a few key takeaways for listeners today?

[00:45:48.420] – Chris
OK, so where where to begin like from from things that we’ve discussed so far? OK, so first let’s let’s begin with the open cloud situation. You have to keep in mind that a cloud is not, by default, open or closed the opposite. It’s not a binary condition. There are like 10000, 10000 to say it’s in between of those two conditions. And in because of that, you have to figure out where exactly does this service belong in this spectrum.

[00:46:23.120] – Chris
Right. And then take one step at a time. Don’t be don’t be planning for this. Don’t be planning for this major migration of everything from mainframes directly to kubernetes take take it one step at a time and in each step, do your homework like investigate what’s what’s available, what can you do better? Learn, adapt. And then this leads me to the final to the final point probably, which is it is a survival of the fittest game.

[00:46:58.800] – Chris
Make sure you have strong a strong team, both on the technical side, but mostly on the softer skills that we that we mentioned, because in the end of the day, this is probably the only way to make sure that your organization will grow stronger over time.

[00:47:18.480] – Ned
Awesome. Yeah, that’s just that’s perfect. That’s exactly that was my key takeaway from from this episode. Absolutely. If folks want to know more about you or mist.io, where should they go? Where can they find you?

[00:47:32.740] – Chris
So they can visit our website, mist.io and they can find me on Twitter as cspsaltis, that’s C-P-S-A-L-T-I-S, OK and you know, go over our blog at mist.io slash blog.

[00:47:50.800] – Ned
OK, awesome. And we will include links into the in the show notes for all that information. Well, Chris Psaltis, thank you so much for being a guest today on Day Two Cloud.

[00:47:59.440] – Chris
Thank you. I had a great time.

[00:48:01.940] – Ned
And hey, listener, a virtual high fives to you for tuning in, if you’ve got suggestions for future shows, you know, we want to hear about that so you can hit either of us up on Twitter at Day Two Cloud show. Or you can fill out the form on my fancy website, Ned in the cloud dot com. Just a little personal promotion here. I’ve got a few new courses coming out on Pluralsight if you’re looking to get the new version of the AWS SysOps admin associate certification.

[00:48:28.640] – Ned
Wow, that’s a mouthful. I have some new courses that will be released at the end of this month. So keep an eye out for those on Pluralsight until next time. Just remember, cloud is what happens while it is making other plans.

Episode 106