00:00
Welcome to the Cables to Clouds podcast.

00:15
Cloud adoption is on the rise and many network infrastructure professionals are being asked to adopt a hybrid approach. As individuals who have already started this journey, we would like to empower those professionals with the tools and the knowledge to bridge the gap. All right. And welcome back to the Cables to Clouds podcast. I'm your this week's host, Tim McConnay. And joining me are our fellow co-hosts, Chris Miles and Alex Perkins. Chris, how you doing? Doing well, man. It's a lovely...

00:43
Monday morning for me here, we're nice and early. Still having my coffee, I'm very sunburnt from my festivities yesterday. I went out and played softball and just forgot to wear sunscreen, man. It was a bad idea. Uh-oh. Yeah, yeah. And Alex, you're getting ready for the Superb Owl, right? Yes, yeah, Chris, since you're in the future, right? You can give us some insights. I can't give it away, I can't give you any results, man. Right?

01:13
Yeah, no, I'm just having a little little family get together. Just, you know, finger foods and kiddos running around screaming while we try to watch football. Yeah, man, it's a weird thing. Like there's there's there's bars here like, hold on. Well, putting on the Super Bowl at 1030 in the morning, like just across the street. And like, who's going to be over there drinking? People take the day off, I'm sure. Day drinkers. That's who's going to be over there. Yeah. We'll be by you soon. What's up?

01:41
Not much. I just finished editing my... I think I'm hoping that it's the last time I have to do any edits on the book I'm writing, so I'm going to get it all patched up and ready to publish pretty soon here. Probably by the time people get to hear this, it'll be out. But yeah, that's what I've been doing all day. Excited, man. Beautiful. Yeah. Awesome. So anyway, so today's topic we want to talk about is the...

02:10
of the myths and and I said myths and legends. There are some legends in the cloud. The legend of the cloud. Anyway, and then you get into like urban legends. Now we need some cloud. Now we gotta do, we gotta go find some cloud urban legends. Like Mothman but in the cloud.

02:35
But anyway, want to talk about some of the kind of the myths that are surrounding cloud, cloud networking, and just kind of debunk the ones that are complete, you know, complete garbage, and honestly give credence to the ones that have some real data, some real insight and, you know, just jump around a bit. So let's just pick one that I hear all the time. And let's see the tear it to pieces or be like, yeah, okay, that's right.

03:04
What do you guys think about the one I hear everywhere that the cloud is someone else's data center? What do you think Alex? Yeah, so obviously while this is true, right? It's probably architected better than yours. That's my thought take, right? It's true that, I mean, pretty much any organization can and probably has in the past, if they're into IT built a data center. But the scale that these CSPs operate on cloud service providers,

03:33
scale that they operate on is just, I mean, it's unheard of except for like the biggest enterprise environments, right? Like I've heard of stock exchanges that have the exact replica data centers. Who else has that? I've never worked at a place that has an entire replicated data center, you know? So all these CSPs, they have so much redundancy and HA and just so much stuff built in safeguards and.

03:59
it's made for customers to consume, right? It's not just made for their internal staff. It's made for everyone to consume. So it's architected differently. I think that's my biggest thing here. Yeah, I mean, it's like the one for one data center. I feel like that is like you said, it rarely exists. Like every time someone's building in the backup data center, it's like, you know, there's cost, there's availability.

04:24
the concerns that come into play and it's like, well, we have to get this switch model. We had to change this thing. It's never one for one like ever. Um, I mean, in some cases it doesn't matter, but other cases it does. It affects your, you know, your automation, your, um, every, every component that you, you put into building that data center. It changes that Ryan. I think, I think the, uh, the, uh, I think you nailed it with, uh, I would say the hyperscaler architecture of the CSPs, you know,

04:51
Yeah, of course it's someone else's data center, but these guys are building at hyperscale. And you know, they're not global. Exactly. And you hear AWS talk about economy of scale all the time about, you know, commodity, you know, how they're able to commoditize the entire thing. So yeah, and of course the orchestration that goes into it is leaps and bounds above, you know, probably the data center you're talking about that's someone else's, you know. Have you guys seen the...

05:20
There's that paper that Facebook published with the pictures of their data center. It's called like F16. We'll add it in the show links, but it's insane. It's like just the scale, that's not even a CSP. So imagine the next step above that, right? We'll add the link. You gotta look through this paper. It's crazy to see the architecture of some of these giant data centers. Yeah. I mean, it's not like...

05:45
Yeah, it surely is someone else's data center. And I kind of touched on this on the previous episode we did about the change in mindset. But like a lot of the a lot of the stuff you typically had to design yourself, like the redundancy, the availability, high availability, those components, a lot of those are just built in. So you don't have to worry about those. And you can focus more on engineering the solution for the application. Right. And it's it's so, you know, there is some truth to this. It is someone else's data center. But it's someone.

06:14
that just builds data centers for the purpose of, you know, this use case, right? So it's like, it's perfect. It's in most cases, it's perfectly built for your use case. You know what I mean? Right. Yeah. And to get to the heart of the whole, the cloud is just someone else's data center moniker out there. It's, you know, it's intended as kind of like to be dismissive of the cloud in general saying, oh, there's nothing special about it. It's just somebody else's data center.

06:41
Yeah, I think it's a little disingenuous because like you just said, Chris, the whole point of the cloud is that, yeah, of course it's someone else's data center, but it's literally purpose-built to do this thing, to be a data center for people. So yeah, so I think we've got that one pretty much not necessarily debunked, but we accept that. Of course it's someone else's data center, but I think we've kind of put the context that matters around it.

07:09
Yeah, it's not even just that it's someone else's data center, but also the way you interact with that data center is completely different. Like, if find me any enterprise data center that has APIs built like that to interact with, you know, everything as a service. Like that is, I mean, yes, it might exist, but the amount of work that it had to go into building that is insane. And many, many large enterprises aren't going to have that capability. So.

07:36
Yeah, it's a trade off, yes, but I think it's worth it. Yeah, exactly. Here's another one that I hear a lot, which is either that the cloud is way more expensive or I hear that the cloud is always cheaper. So it's one or the other. How do you guys hear it usually? Yeah, yeah, Chris, you wanna start us off on this one? Sure. Yeah, I mean, as it goes with pretty much any situation in IT, it's always that common answer, right, of it depends, right?

08:06
I mean, if you do, if you do a lift and shift, yeah, it's, it, you, you bet your ass it's going to be more expensive because that's, that's not how things are meant to be run. But I think even because, you know, obviously the show is focused on, on the networking piece of it, right? So I feel like the lift and shift mindset is always focused on the application, which is very true. But I think that's very true for the networking component as well. Like if you think that you need.

08:34
a certain amount of networking constructs put in. It's like maybe you can simplify it quite a bit and pay much less for the services that you need to run. But sometimes you don't, sometimes you need to add more. Maybe it all fits into a nice box for you and when you put it in the cloud, it separates. So, sadly, it depends. I know that's the...

08:58
Kind of the top out answer, but you know, it's, it's, it's prevalent in this one. So how do you feel Alex? It's not a cop out though, because it really does. Yeah. I mean, just to add a little bit, you know, I, it's really bent on it's me, like, yeah, how the cloud is consumed, right? Because a lot of this started with shadow IT. So of course it's going to be more expensive because the people that are using it, that started using it, or just.

09:26
doing whatever they wanted just to get their app up and running. They're not caring about, Oh, this type of infrastructure costs more, or we don't really need like, like a crystal sand, you know, the network, maybe you don't need everything to be connected. Maybe you do. And there is even more cost, but there's plenty of changes where it's like, Oh, we just added all these extra constructs into every VPC that we created for no reason. It's just the sprawl that just comes out of hand so quickly.

09:54
Yeah, I've seen so many customers where like, they can't, they have, yeah, I've seen so many customers where like, they think like, okay, everything has to be segmented. So like, they're putting like 10 VPC endpoints in every VPC. And it's like, do you pay for those on an hourly basis? Like you've shot yourself in the foot. Yeah. Nat gateways everywhere. Yeah. For everything, right? Every VPC's got a Nat gateway. Yeah. Absolutely. Especially around cloud architecture. I think the whole idea of,

10:24
of it being more expensive or not as expensive as in absolutely. Yeah. So just probably for the sake of the listener, should we expand on what shadow IT is just that some people may not be familiar with the term? Yeah. Um, yeah. I mean, to me, I think everyone's gonna have a little bit different definition, but shadow IT really came about when the cloud first came out. We, I think we touched on this in like the first or second episode. It's.

10:50
It really started getting big because devs, it's so easy. You just have a credit card, sign up an account, add a credit card, and then you can spin up whatever you want and it's all just gets charged to the corporate card. So the shadow IT is basically IT that was going on in the background. That wasn't, didn't fall under like the framework of the organization.

11:11
Yep, going around the process completely, not tracked, not part of the process. Yeah, absolutely. Suddenly the finance department's like, what's this bill that we just got for AWS? What's that? Right. Yeah, I mean, it's like they didn't want to wait, you know, six weeks for someone to spin up a VLAN. Right. Yeah, so. Yeah, we're not even saying it's right or wrong. Right. It's just, yeah. That's why it's called shadow IT, because there's a lot of, there's credit cards being run and the governance and the process is not.

11:38
is being followed or is not being followed. And therefore it's kind of like see sneaky around the back and nobody knows what's going on. Yeah. I wanted to add a little bit sneaky. It's sneaker. Dude, remember sneaker, sneaker net. That's where you take the USB USB drive and plug it into the different computers.

12:02
Now, I wanted to add a little bit before we move on from this one, which is, first of all, I like that you related it specifically to cloud networking, Chris, because you're right. And this is a trap that a lot of us moving from network to cloud are going to fall into, which is we want to, you know, everything's already so different. We want to use something that's kind of normal for us or that we're familiar with, right? So the...

12:28
the temptation to spin up an NVA, a network virtualized appliance router or firewall or whatever everywhere and use them in the way that we know how to use them is powerful, right? You're like, hey, I know how to build IPsec tunnels with my CSR or whatever, my Juniper equivalent. And I don't have to know how to do all these weird cloud constructs, but the truth is that you can be panted through the nose. You don't want the X-Line fabric in your cloud.

12:57
Well, I mean, I do, of course, but the business might have different feelings about that, right? So I think that's a really good call out specifically. We normally speak in terms of the app, but the truth is it's more like a, it's almost a holistic whole body view to say that it's more expensive or less expensive as a whole based on what we're doing, you know, whether we're...

13:23
building our apps that way, being over consumptive with our apps, building extra constructs, or lift and shift is definitely the worst thing in terms of money. Then starting and refactoring and decoupling and using PaaS services and using the services that are available. So lift and shift. I mean, how could it possibly be cheaper to lift your workload, changing nothing into the cloud, and you've just put a middleman.

13:49
to take the money between you and your workloads, right? Like how could that possibly be cheaper? But one thing we didn't talk about is how could, you know, is it's always more expensive? No, it can be cheaper, right? If you're using serverless architectures, you know, if you could downsize your hosts, the amount of VM hosts you're running in your data center in a way that, and consume serverless, you know, for micro pennies on the dollar, you know, for functions and stuff, you know, there is a way.

14:17
to make the cloud cheaper. It's just depends on what you're doing, the workloads and how ephemeral they are and how long they have to live, static workloads don't work as well. So anyway, I think, is that a good stop point? You guys wanna move on to the next one or do we have more to say about this one? Yeah, I mean, a lot of these tie together.

14:36
Yeah, well, just one more point to add on that. Like I think the biggest case is familiarity, right? Like you don't want to switch to using some new thing. You want to know what you're comfortable with. And it's very common. The thing you're comfortable with might be, you know, a way more expensive than the native construct that you can use to do the same thing, right? So, cause you're paying, you know, you're paying instance costs for whatever the instances you're running. You're paying licensing costs to the MVA vendor. You're paying, they may be charging per the way you're using it.

15:05
on a subscription model. So it's like, you can, you gotta change things up a little bit if you want to make it cheaper. I think that's, like I said, it's very app focused. Like, hey, when you move to the cloud, you gotta change the app. You gotta refactor the app to be cheaper, but like, you gotta do the same with the infrastructure too. Like you gotta use the components that make sense for the use case. Yeah, and it's, you know, again, it's not just lift and shift of the app.

15:33
Like you're saying, it's lift and shift of the infrastructure. You can't do it in the same mindset. Yeah, that data center mindset. Yeah. You can't just replicate your data center in the cloud and that be cheaper, right? It's not the, it's not going to happen. Okay. So let's see, let's pick a different one here. Well, actually this one, we'll come back to this one because we kind of already covered it, so let me pick another one from our list. I got a list that we put together before the show and I just want to.

16:02
jump around and pick the next one based on how the flows go on here. So here's a good one. So the cloud should be used for everything. These are the ones and I don't know, you guys probably remember this. I remember being, I remember very well when I was an enterprise network engineer, I remember we got a new CIO and we had just built a new headquarters and built a brand new data center and populated it and moved everything and whatever.

16:28
brand new data center, hot aisles and cool aisles and all that stuff. And the CIO we just hired walks in and is like, I wish I'd gotten here six months ago. I could have kept you from making this million dollar mistake. He was 100% cloud. He was gung-ho. Everything should be in the cloud. He knew nothing about our workloads, about how our workloads worked, how our apps were built. What...

16:55
It was just, it didn't matter, right? Like, Hey, it costs a million dollars to build a data center. You know, we could obviously do everything cheaper in the cloud. And, um, and he was very much like clouds we use for everything. So I don't know if you guys have run into that in your, you know, where you come from as well, but that was that, that one left an impression on me. Yeah. I mean, there's, and it's, it's kind of, it kind of sucks because a lot of this, like you said, you were giving the example of a CIO, a lot of this does come from like C-suite people that you think.

17:24
should know better, should be taking a look back. And you mentioned it perfectly. You said the workflow, right? They didn't understand your workflows. There are workflows that are way better not using the cloud. I mean, like vertical scaling workloads, right? Things that are always on 24 seven. The cloud gives you certain benefits. And if your app doesn't take advantage of those benefits, you're just paying more. Yeah, that's where you're gonna run into these negative. It's just, yeah, you really.

17:52
Someone needs to take a step back and look at the workflows. Yeah. I mean, I think we've, we've, we've kind of pounded on this on the, on the last episode as well, but like the hybrid is always going to be a thing. I don't think there's ever going to be a situation where everything's running in the cloud. Um, I mean, there's, there's always going to be people, I mean, there are definitely companies that started recently, probably in the last, you know, five years or so that, that run all completely.

18:17
Cloud-based infrastructure, SaaS services, like nothing's running on-prem, everyone's remote. That's great, but I don't think that's ever gonna be the case for large enterprises and things like that. Everything's gonna have an on-prem component. And honestly, I've never worked in enterprise, so I haven't seen a lot of that where someone comes in and wants to change everything just for the sake of changing it to, you know, to buzzword A.

18:44
that they heard on, you know, luckily for me, I'm usually on the other end of it. I'm the person they paid to come in and help with whatever that transition is. I have to make it sound like it's really nice. But you know, it's, but yeah, I think it's, you know, you got to do your due diligence in those situations as well to be like, you know what? This doesn't make sense. Like you're gonna, you're gonna end up paying more. You're not going to be happy with the results if you do this.

19:14
moving this workload to the cloud or whatever it is, you know, so, yeah, it's not a blanket thing. Absolutely not. Yeah, it definitely can't be, right? Because actually, using the story I just told as a perfect example, right? This company I was with had just spent a million dollars building that data center, right? So in what possible case would it make sense to have abandoned it all and

19:41
you know, to move the workload, right? So yeah, I mean, we already have the infrastructure at that point. So there are definitely customers, sorry, customers. There are definitely enterprises out there that are full cloud. Sorry, I work for Aviatrix. So I work with customers who are often enterprises.

20:02
So, you know, who have, who are definitely embracing that idea that, Hey, we, we've been, we have this mandate to get out of our data centers because we don't want to own infrastructure. We don't want to own real estate. We don't want to pay rent, you know, the cooling heating, like there, there's definitely cases where that makes sense. And usually it's when the data center that you've built is old and you don't want to do the upgrades, you know, you don't want to build it's a big one. Right. Why not?

20:31
Right. Yeah. But there has to be a use case, I guess, is the point I'm trying to make it. There has to be it has to be right. And it has to go beyond this this blanket idea that, you know, we're getting out of data centers or moving to the cloud like, you know, tablets and Moses from from on high. So it just has to make sense. Yeah, you shouldn't have to force it to make sense. Yeah, exactly. The math should be the math. There's a saying, I think, in like the VC space, like the venture capital space,

21:00
Like you're crazy if you don't start in the cloud and then you're crazy if you stay in the cloud. Once you start out, you're going to have that growth. But once that's steady, it doesn't make sense anymore. And that's the kind of workloads that aren't a great fit anymore. Starting the cloud is great, especially if you're in a company, if you're in like a startup that doesn't have a data center, doesn't have infrastructure.

21:22
Like, why wouldn't you, why would you build that before, when you're in startup mode, and it makes no sense at all. Startup mode's about the MVP, right? It's a minimum viable product. Exactly. And it's all that, it's deaths. They make the product, you don't need all the extra infrastructure pieces. Right, it's when you reach the MVP, and you've got customers, and now you need criticality, now you need critical, you know, critical workloads, critical, you got an SLA to keep, and all of that, right? Yeah, the SLA, yep. That's a big turning point. Yeah.

21:51
So anyway, any other final thoughts on this or I'll move on to the next one? I think it was beat to death. Let's grab this. Right. Actually, let's, let's, let's segue right into this next one, which is, multi-cloud doesn't make sense for any enterprise. And now you'll find that a lot of these myths tend to be,

22:13
declarative like in all cases, like for example, it's never, it's always cheaper in the cloud or it's never right to do this. This one, you know, multi-cloud never makes sense for any enterprise. And of course, it's not true because there's always going to be a case where it makes sense. So Chris, why don't you lead us off on this one? Sure. Yeah. I think this one kind of just piggybacks off the shadow IT thing. It's like.

22:37
Yeah, like once a dev learns that, you know, something runs better in GCP or, you know, has this feature that really applies to them, like they're going to go stand up in GCP. And like, before you know it, you're running a multi-cloud environment, right? Maybe you guys want to run on AWS, but all it takes is that one thing to transition over and you need to find a way to bridge those things together. Right. So it's, you know, I mean, that's one way it happens, but also

23:06
I think enterprises need to, you know, they make the decisions that, you know, we need to run things where they run best, right? And I know that's kind of a Tim and I work in Naviatrix and that's kind of a foundational piece for us is we're very focused on multi cloud. And you know, I think with, you know, we see a lot of customers that have that focus because they want to run the workloads where they run best. You know, Azure AD is a great component to add to anything and you know, more and more

23:34
I think, I feel like the cloud. Yeah. The CSPs are, I mean, they don't, they don't really want to work together. They don't want to, they don't want to give you a reason to, to stand up things in another CSP, but they are kind of adding little things here and there to let you do it, or at least, you know, bridge. I feel like there was a recent announcement. What was it from OCI and Azure that they were starting? Yeah.

24:02
I think that's probably a little bit more focused on getting more of the market share for OCI, but you know, it's still it's a step in the right direction, right? So yeah, I think you don't necessarily choose to be multi cloud. The decision kind of gets made for you. But yeah, I don't know how you guys feel. What? I'll go next. What I see a lot in this one, especially as pertains to being a

24:27
a myth is that the myth that goes with this one all the time is that enterprises only do multi-cloud for redundancy purposes. I don't know if you guys have seen this or heard this idea. And then of course, everybody else is like, that's the dumbest thing ever. And I tend to agree actually, if you're doing multi-cloud only to diversify.

24:49
Because think about what that means, right? You're double building your infrastructure in two different clouds. Like that's insane. Like that doesn't make sense. And I agree with, if that's what they're saying, then yeah, I would totally agree with that. It doesn't make sense. But by itself, multi-cloud makes sense for a lot of enterprises. Like you said, Azure AD, you know, a lot of customers, hell, most enterprise customers probably have some form of Active Directory on-prem and they want to use AD Connect and Azure AD

25:18
You know, it makes sense to integrate in Azure with AD, you know, but they might be running container workloads and GKE maybe, you know, is offering a lot nicer options than, you know, they don't want to do it in AKS or by GKE, I mean, I'm talking about the Google Kubernetes environment in GCC, in Google cloud.

25:40
versus say like Amazon Kubernetes services or whatever. All the clouds are kind of differentiated a little bit. Not a lot, they all try to offer, like you said, Chris, you kind of, they don't want to give you a reason to go anywhere else. But there are definitely cases where the cloud runs, certain things run better in different clouds. So I think in that respect, we're seeing that already, especially Chris and I at Aventrix. And then, but yeah, anyway, sorry. So to get back to it, the redundancy thing, yeah, totally debunked.

26:10
be absolutely stupid to do multi-cloud redundancy, like as a redundancy strategy. That's dumb. So what do you think, Alex? Yeah, I had a couple of points. You guys brought up some things that made me have even more points. So I'm gonna start with just answering the question, then I'll get to the other points. So I think, so for me, coming from the enterprise side, I work a lot on, now I work for VAR, so I see a lot of different enterprises, but.

26:39
I do a lot on governance frameworks. And this is the myth that kind of goes with this that I always hear. It's like, well, if you have a governance framework, everyone uses the same cloud. It is never true in practice. It never works that way. You can start with all the good intentions in the world, but at some point, um, and this is just a point I just wrote down, like SaaS applications, if they're hosted, if one's hosted in Azure and one's hosted in GCP.

27:05
your multi-cloud, right? Like it's, you know, you still have to find ways to connect everyone to all these different clouds. And it's this governance framework always ends up evolving into a multi-cloud governance framework. It's never just a single cloud governance framework. It may not be your specific dependency, but it is a dependency, right? Like you, you inherit that dependency for whoever the SaaS provider is, right? Right. And if you don't, I mean,

27:31
People are going to use, like you said in the beginning, they're going to use the tool that they are most comfortable with, that they feel is the best, most advantageous. If you restrict that, good luck. They'll just go around it with a shadow IT, right? Yeah. And then I just wanted to add the quote that everyone loves is, every enterprise is one acquisition away from multi-cloud. Oh, I love it. That's so true. So true, man. So true. Everyone's like, well, we don't acquire anyone. Okay. Well, we'll see what happens.

28:00
Yeah, no, I mean, we deal with that a lot. I would say I would say more than I mean, obviously in the vendor space we're in, obviously we deal with it an awful lot and mergers and acquisitions is a huge reason why enterprises that are, you know, I say enterprises are companies and go in multi cloud, you know, they inherit the multi cloud environments. Yeah. And, you know, on the Kubernetes stuff you guys were talking about, almost all of I think, at least the main three, Google, Azure and AWS.

28:30
They have like an anywhere kind of version. Oh, okay. It doesn't seem like that's how a lot of people are using them. It's more like when this person is in this cloud, they're using that cloud's Kubernetes service. Even if they're in multiple clouds, right? They're not expanding like GKE anywhere to AWS and Azure and GCP. Interesting. Yeah. That's really good, guys. Anymore thoughts on this one or do you want to move on? That's all I have. Yeah, that's all I got. Okay.

28:59
Okay, I like that. Okay, I want to hit this one because I want to make sure we have time for this one because this one is we talked about this a lot before the episode. We are too small to take advantage of the cloud. I mean, this one. Yeah, I, this one is insane to me. Like, we just talked about it with startup, with startup culture, right? Like, you know, how could you possibly be too small to take advantage of the cloud? You know, if you are small, then let's think of like,

29:27
know, Pop's donut shop or something like that, right? You know, they're thinking, you know, Pop is thinking, man, I, what do I need the cloud for? I'm way too small for to worry about that. And, you know, the truth is that, um, of course, you know, Pop doesn't have any infrastructure, so think about running a website out of S3, which is the, you know, simple, uh, the storage in AWS, right? They could run, he could run his, uh, if it was just a serving static content, he could run the whole thing out of an S3 bucket.

29:53
you know, or do a serverless for taking orders, you know, on the internet. There's a thousand ways that, you know, pops donut shop could leverage the cloud and should be leveraging the cloud rather than worrying about, you know, having to keep a computer sitting in his donut shop, uh, connected to a router serving a webpage or some, some weird other way to do it. Um, so I think that one, to me, that's like not only debunked, but it's almost like opposite land stuff. So what do you, what do you think Chris? Yeah. I mean.

30:22
I feel like it's complicated because there definitely are some use cases where it makes sense. I guess in my mind, when I hear take advantage of the cloud, I'm thinking, you know, running applications, you know, kind of full scale, but you made a good point there with the S3 website. But like, you know, it's like, I guess to what extent are you consuming it? Are you really taking advantage of it? Like you're running a static website. That's great. But like, would you really put them in the, I mean, that's a very...

30:51
very small minority of companies using the cloud for that. Oh, certainly. Yeah, but it's, I mean, I feel like in, but in a lot of cases, if we are talking about migrating applications, we're talking about changing things like that, I feel like there is some truth to it sometimes. A lot of companies are so small and they don't have the expertise, the infrastructure, the availability, the engineering to refactor those things and move them over, right? So it's, you know, I mean- That's a good point.

31:19
It's because it's like, I feel like it's easy for us to be like, you need to refactor, you need to do this, you need to do that. When, you know, the fact of the matter is, you know, you got to have people to do that. You got to have people that know how to do, you got to give them the time to train, you got to give them in it. And for small teams, it could be a lot of apps you need to move. That means a lot of apps you got to refactor. A lot of infrastructure people that need to completely retune what they're doing. And you know, it may be small, small data centers.

31:48
Um, so, you know, I feel, I feel like there's a bit of truth to it. Well, it's okay. And I'll, I want to hear what Alex has to think too, but I'm so, so is that a size thing? Well, I guess they kind of go high as hand in hand. The idea is that, you know, if you're small, you probably don't have that expertise, right? It really is an expertise problem, but I get where you're coming from. That it's, you know, if you're small, you probably don't, you're probably not rolling with the expertise. So, yeah. Yeah. It's about like, if you're small.

32:14
You have a business to run, right? You got to, you got to, you got to do what you got to do to let the business run. And maybe you don't have time for these other activities, right? Or, and then, you know, if you want to bring someone else in who can help with that, that's money. So it's, you know, it's, it's kind of a Robin Peter to pay Paul type thing, right? Right. Yeah. I mean, I initially was going to go with him a lot on this, but I think Chris, Chris made a really good point. Uh, you know, I mean, I live in a really small town.

32:43
There's not a lot of IT people out here, right? So your pool, especially for small towns and small shops, they don't know, they don't even know what the cloud is, right? Like how many conversations have you had with people that have no idea what you're talking about? That's true. So, I mean, I get it from that point. And yeah, maybe it is just a size thing. Cause on the opposite end, we talked about this, I think episode one or two, but having a CDN.

33:11
available at your fingertip, content distribution network, a global CDN at your fingertips. Right? If you're like, I don't know, you're setting up like a e-commerce shop out of your house, you suddenly have global reach. It's insane to not take advantage of that immediately instead of trying to, how do you do that yourself? Right? Like that's never gonna, the scale there. So again, it comes back to it depends like it always does. That's true. I mean, and, and, you know, I would say that the clouds make it

33:41
I won't say that they make it as easy as possible for you to consume, you know, to, to like look at AWS light sale where you can get a website up and running in like what five, like a minute or so, right. And you can connect to the CDN. So I'm with, I'm at kind of half and a half. I started off very strongly in favor, but I do appreciate what you're saying, Chris, and I do agree that that is just as relevant. So I'm kind of half and half on this one. Actually. Yeah. I like this one.

34:11
Yeah, so, you know, whoever's listening out there, I personally would love to hear your opinion and specifically, I'd love to hear if this is something that you're either struggling with or have run into or done anything with yourself. So you know, for the, so, so mom, mom, if you ran into this, I know you're listening. No, I don't really. Okay, so let's see. We're running out of time here. Let's get one more in. Okay.

34:39
Uh, let's see. So, oh, Alex, actually I want to hand this over to you because you specifically wanted to talk about the, the, the AWS is seven hours. And so I kind of want to give you the platform to talk about that. Cause it's, it's relevant here, I think. Yeah. And, uh, this is going to be interesting. So the seven hours, I believe they've added to it since I, so it's the eight hours now, four, no, so now I thought it was when I went to look this up.

35:05
Was it the four pillars? Not the pillars, it's not like the one with the protective stuff. The framework? Yeah. Did they actually refer to it as the four Rs? Was it called the four Rs at one point? No, it was originally. Well, because now it's seven Rs. So it was four Rs. Yeah. That's just our phrase to say four Rs. It's not hard to learn.

35:30
AWS is four Rs. In the show nuts, because there's actually a really cool, uh, diagram that it shows kind of like a workflow. And there's even a link that gives you like a draw.io template to follow so that you can kind of determine what you would do with an app when you're considering what to move it or not. That's really cool. Yeah, it's awesome. And so basically I'll just lay them out. They're refactor.

36:00
replatform, repurchase, rehost, relocate, retain, and retire. Right, some of these are self-explanatory, retire, you don't do anything with it. It's just gonna either be replaced or it's just gonna run out, right, get decommissioned. But I wanted to just go into some of these because a lot of these are things you gotta think about. This really ties into like the lift and shift and we're done. Like you don't really wanna do that.

36:27
There's things you got to think about, like can you refactor? Can you put things into containers and microservices? Can you re-platform, right? Instead of just keeping it in like a rel, you know, a rel distribution that you have, do you move it into some kind of AMI that's provided? And AMI is, what is it? Amazon machine image. Yeah, it's just like an image for an OS image, right? And then there's repurchase, right? This is like buying things from the marketplace. Off the shelf. Yeah.

36:56
independent software vendor, you'll hear it called ISVs. These are all your NVAs that Tim's talking about, right? Stuff that's third party that you can buy as a service instead of rolling your own. It's just a really cool way to look at it. The flow chart, I highly recommend looking through it and getting an idea of the things that you should think about when you're trying to consider these points. Yeah, so you bring it up and we talk about what the Rs are, what they are, but I guess let's work our way back. Let's say what...

37:26
What, what myths or industry kind of comment do you relate this to? To me, this was really about like the lift and shifting we're done, right? Everyone thinks, I'll just put my app in the cloud and that's it. No, there's seven Rs just, just from AWS that they came out with. Think about all these things. You should see the flow chart. It's not insanely hard to follow, but it's, it's significant, right? You have to really think about all the options. It's one of those like.

37:53
refactoring this code, yes or no, follow this line. She's gonna adventure with your app, right? Turn to page 35. Yes. It's funny because even the post is like, it's like, it depends. This doesn't apply to everything. So it depends what the workload is. Right. You gotta decide what to pick, right? And of course it's biased towards AWS, right? So there's other services to think about in other clouds. Yeah. I mean, it's, but it is, it's kind of a more,

38:23
holistic construct, right? Like it doesn't directly, I mean, it's obviously for, it's gated towards AWS, but it can be applied to more than just that. I mean, they even have like, I'm surprised we didn't talk about this earlier when we were talking about the moving data centers. Like even when you think about the retain option, maybe that is more focused on the hybrid piece of it, right? You keep it on-prem. And that's like, we were talking about moving the data centers and making the data centers

38:52
you know, building a data center that you can't build it like it, like one of the hyperscalers do. Right. And that's why something like AWS Outposts even exists. Right. So that you can take all the benefits of what they've done and they hand it to you in a physical form factor that you just deploy and you have that compute running at the edge and you interact with it just like it's, you know, just like it's a different region in your AWS construct. And like that is a, that is a like.

39:18
component where it's like you get all those benefits and you still have your, you know, your physical component. If you, if you're, you know, latency sensitive type things like that, you know? So yeah. Yeah. That's the local zones and you know, this is wavelength. There's all these like things, all these products that are coming out with and bring it, yeah. Push the cloud out to the edge. Edge compute is definitely an interesting new way that the clouds are, I mean, it makes sense though. They have to.

39:45
Like, because at the end of the day, they're not going to build a data center in every town, right? So how do they bring it closer to you? They're going to use wavelength and local zones and outposts. And I know we're using AWS terminology here. You know, I'm not as good with, I'll be honest, I'm not as good with Azure's names. I'm okay with Azure, but I'm very bad with names. But like, you know, Azure has a lot of the same stuff. So does GCP. All the CSPs are doing this. And for the record, anyone wondering, wavelength is kind of like,

40:14
cellular pops, if you will, like 5G. Yeah, it's 5G. Yeah. They'll bring basically AWS will bring 5G to your doorstep. Yeah. Private cellular network. Yeah. I mean, I feel like it's like the CSPs are the first ones that are really leveraging the advantages of 5G, right? They're probably going to be the biggest. Yeah. Right. Yeah, for sure. Because doesn't a 5G have like a.

40:41
a lot of, and this is getting outside. I know it's network still, but cellular's not my strong suit. But I think one of the big things about 5G was like the really high speeds that you need are distance locked, right? There's like, you have to be within a very short period of distance. Yeah, 5G brings faster speeds, but at a closer distance. Yeah. That's why it's really good for like factories, IOT type AI solutions for factories and stuff like that. Right. So. All right, well.

41:11
Dude, we have so many more of these that we could roll through, but we are, we are pretty much out of time for today. So I hate it because there's some really good ones in here. Maybe we'll do another episode down the line. We'll do a part two once we add a couple more and go through them all. More myths will come up. And it'll change over time. We'll get more of the gents, we'll get Muff Man. Yeah. So, but also if you're listening to this and there's more that we didn't cover that you want, that you want us to talk about or.

41:39
if you maybe want to come on here and talk about something. Hit us up, let us know. We'll see what we can do. Yeah, absolutely. All right, guys, well, this is a bit of a blast. Let's do it again sometime. Yeah. All right. Maybe in two weeks. Yeah. Maybe, maybe in two weeks. All right. Okay, well, this has been Cable to Clouds, and Cables to Clouds, rather, and thanks for joining us today. Take care. Cheers. Hi, everyone, it's Alex, and this has been the Cables to Clouds podcast.

42:08
Thanks for tuning in today. If you enjoyed our show, please subscribe to us in your favorite podcatcher, as well as subscribe and turn on notifications for our YouTube channel to be notified of all of our new episodes. Follow us on socials at Cable's 2 Clouds. You can also visit our website for all of the show notes at cables2clouds.com. Thanks again for listening and see you next time.