Future Proof Your Network Career in the Cloud Era with Du'an Lightfoot - C2C025

Chris Miles: 0:00

Welcome to the Cables to Clouds podcast. 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.

Tim McConnaughy: 0:29

Hi everybody and welcome back to another episode of Cables to Clouds. This is Tim, your host this week, and we have a very special guest with us this week. Duane Lightfoot is here from AWS and actually it's funny to introduce you like this. We've known each other for a while now but hey, man, it's finally good to be able to collaborate and get you in and do something.

Du'an Lightfoot: 0:51

Hey, team, thanks for having me on. It's good to come connect with you, tim. Like you said, we've been friends for a while. We always talked about doing something, so it's good to collaborate on something that's going to help the community.

Tim McConnaughy: 1:06

Yeah, definitely, that's what we're all about here. Hey, dude, we just came off of a re-invent. We're going to do a movie, we're going to do an episode about re-invent here in a little while. But I mean, duane, you got to deliver multiple talks at re-invent so I mean just let's take a couple of minutes. We'll get into kind of you and your journey and then we're going to talk about our skills gap in networking tonight. But I would love to know just upfront, while it's fresh in mind how was re-invent?

Du'an Lightfoot: 1:37

Hey, re-invent was great. This is my second one. I presented last year this one. What I found interesting was the excitement around cloud networking. We hosted a networking game day. This is a brand new launch for AWS. The team worked really hard to put this together. We had about 250 people in the room, so the room was over like 80% full and people were just excited to dive into cloud networking. It was just amazing to see the excitement, the energy and just the people that came out. You know what I mean, because oftentimes when you talk about networking, it's one of those things that nobody talks about until something breaks or they need a new VPC or they need something new, right. So just to see the engagement from the community, not just in the game day but also in the other sessions that we had going on throughout the event.

Tim McConnaughy: 2:25

That's great man. I'm glad to hear it. I mean, obviously, cables to clouds here. Right, we're all about the cloud networking, so it's nice to see cloud networking get more of a seat at the table, a little bit more of the spotlight. Okay, so sorry, I just had to ask that real quick. But for the people who don't, for the two of you that might be listening to this that don't already know, dwan Dwan, if you wouldn't mind, take a minute, introduce yourself, tell us what you do over at AWS.

Du'an Lightfoot: 2:51

Yeah, so my name is Dwan Lightfoot. I'm a developer advocate at AWS, focused on cloud networking. I started my journey a long time ago right what? 20 years ago? Nit and Air Force, doing Windows NT to Windows 2000. Basically on PDCs, the PDCs to the introduction of Active Directory. So I kind of seen technology evolve throughout the years, from those physical servers to the virtualization, to the containerization, now to the cloud, and it's been an interesting ride. And it's been great to not only see the application side evolve but also the networking side, because now we're doing some really cool things in application networking, we're doing automation. All of these things are starting to come together to make the network engineering career look a lot more interesting and just kind of refreshing. You know what I?

Chris Miles: 3:40

mean and I think maybe some listeners might hear that and hear developer advocate for networking, and those two things don't relate right. So how are you bridging that gap and doing kind of that developer advocacy with a networking standpoint?

Du'an Lightfoot: 3:57

Yeah.

Du'an Lightfoot: 3:58

So I think, as far as being an advocate for networking, I probably started doing that in 2018, when I started my YouTube channel and I started talking about Cisco certifications, router switches, how to learn networking and things like that.

Du'an Lightfoot: 4:10

And then, around the same time, I kind of shifted to doing Python and talking about network automation, which kind of led me to DevNet, and so those of you who have heard of Cisco DevNet on that journey I was doing network automation, I created Python for Network Engineers course that you can find on Ciscocom and training and all that, and while doing that, they just kind of the natural progression as you learn APIs, as you learn automation is to eventually start using infrastructure as code and then eventually make that migration to the cloud, depending on your organization, depending on your goals.

Du'an Lightfoot: 4:46

But oftentimes there's either going to be an opportunity for you, as a network engineer, to apply those to the cloud or there's going to be some type of cloud network engineering team that you're helping out, assisting, standing up VPNs, and you're finding new ways to do that by using infrastructure as code. So, to make the long story short, on my DevNet journey, it kind of led me to the cloud, which got me here to AWS, since I understand networking, I understand software development and now I was just kind of bridging the gap when it comes to communicating with not only developers, but also network engineers and the service team, business stakeholders as a whole.

Chris Miles: 5:20

That's really cool. I think we've talked about on this show, about how you know, sometimes on-prem networking and automation kind of feels like a bolt-on thing that was added later, whereas in the cloud it was kind of built from the beginning, right. So it's just it's good to hear you say that you know you just naturally make that progression to the cloud just because of you know, ease of operation and things like that.

Du'an Lightfoot: 5:42

That's a great point, chris, because in the cloud, when you're in, like the AWS console right and you click on the VPC console, you click on the EC2 consoles all of those are API calls. They have a front-end. Every service connects and builds on top of every other service on AWS. All of those are API calls and so if you're using API calls, you have structured data, you can use declarative code to communicate with your infrastructure, deploy your infrastructure, like. It's just a whole different mindset as compared to on-prem, which then not saying it's a big hurdle, but it's actually just a more structured approach on how we deploy and how we automate and could build process around the upgrades, the change windows and all of these things that we do on-prem.

Alex Perkins: 6:23

Yeah, and this is something we've been talking about a lot lately is kind of bringing those models, like the cloud models, to on-prem and how that gets built out and like you see it right, like you said, it started out with a lot of the DevNet stuff where it was kind of bolt on APIs right, you had like NetCon Frescon but you start seeing all these like fabric overlays that are more meant to provide like that API, like more cloud-like approach, right. So it's. Yeah, it's like you said, it's been cool to watch just kind of that transition through the on-prem networking portion of things too.

Tim McConnaughy: 6:56

Yeah, I mean I'm so glad we have you for this episode because I mean we want to talk about. We came to talk about was kind of the gap, not only just the gap in skills between network engineering and, say, cloud network engineering, or really any two technologies or any three technologies, but how you're bridging that divide, how people are learning, going from, say, network engineering to cloud network engineering, or not even just going from, because it implies that you're leaving one behind and picking the other one up or going to the other, but really you're just adding. You're adding to your skill set and to hear your progression career-wise, it's exactly what we're here to talk about, right, which is building on it from soup to nuts up to where you are now.

Du'an Lightfoot: 7:50

Yeah, it's building blocks. I mean, of course, networking at the foundational level is something that we need to understand. Although, in the cloud, a lot of us are abstracted at that layer, one, layer, two level, we still have to understand that to even connect to the cloud at the on-premises level. So, understanding that, having a foundation and networking, and then, before you even dive into the networking piece in the cloud, understanding the building blocks, right, understanding compute what does compute look like in the cloud? What does storage look like in the cloud? What does security look like in the cloud? What does networking look like in the cloud? Understanding all of these fundamental building blocks and what the cloud is as a whole okay, how does it look like to protect in the cloud? How does it protect the light to secure connectivity to the cloud in the cloud itself. And then, what are the, I would say, abstraction layers, like the Amazon VPC? Okay, that's one compute type. What about serverless? How does that actually look? And then understand okay, how do I connect to services? Oh, there's public services on AWS. Oh, wait a minute, ipv6 is really a thing in the cloud. So now, what's understanding that IPv6 is something that you probably need to learn If you've been avoiding it at this point. I don't think you can do it too much longer. Now it's time to really dive into the IPv6 and understand it. Okay, what services support it? How do I secure it? Oh, I have a router table. It has IPv4 and IPv6. I need to update that. What about security groups? We're running IPv4 and IPv6. We have to update both as well.

Du'an Lightfoot: 9:22

So understanding how to apply these principles in the cloud is, I think, part of the journey. But the other part of the journey since we're talking about the networking tools, the fundamental tools, and then understanding the cloud the other piece is going to be the tools themselves. How do you give visibility Right? Because it's different. Running a ping isn't always going to work, especially if you have security groups who apply. So one of the tools you can use in the cloud to get the visibility from source to destination, because it's going to look different. What if you're running different clouds? How do you get visibility between the two? How do you trace in between the two? How do you provide that secure connectivity between the two? And then the next piece is going to be BGP. I think when it comes to routing protocols it's a little simpler, but you still have to understand it. It's something that, if you have not been using it on-prem, if you're connected to the cloud, the GP is probably something you're going to be running.

Tim McConnaughy: 10:16

It's funny to yeah, no, absolutely. And I'm forgetting who we were talking with offline. It might have been Andy, actually from the Art of Network Engineering crew, where he was like I've been railing against network, not to put words in his mouth, he could speak for himself, but he was something. I believe he said something along the lines of I've been railing against network automation for so long and I just now realized that BGP is network automation to some degree. Anything dynamic, anything really dynamic Not much, right, ralph. Anything dynamic right Learning routes dynamically rather than having to statically plug them into the CLI, could be thought of as some form of I mean, it's very basic automation. We almost don't even think of it that way anymore, but really so yeah.

Du'an Lightfoot: 11:00

Yeah, yeah, yeah, yeah, yeah, yeah, yeah, yeah, yeah, yeah. I would have been fortunate to work with a lot of architects. That's been around a long time. So when I first started talking about network automation in my organization, to them it was nothing new. They've been writing porous scripts, they've been using EEM, they've been doing all of these things forever. So to them it's like, yeah, it's just something that was glittery. We've been doing this a long time. It's nothing new. I really didn't understand it until I really started taking a step back, for on-prem this isn't really anything new. But if you get into the cloud, that's what things kind of look different and they kind of change.

Chris Miles: 11:34

Yeah, and just to circle back to what you said about the building blocks, I feel like that's obviously that's where you got to start every time. But if we think about, like ourselves, like the four of us on this podcast today, that we have close to a collective 80 years of experience in networking, now there's a lot of people getting into this industry now for the first time and they're naturally drawn to, it seems, like two things that are very popular cybersecurity and cloud.

Chris Miles: 12:04

The conversation has changed to that. What do I default to? But, as someone getting into cloud, where do they get those building blocks from, whereas we got it from Cisco for years and years? Are you seeing people getting in the cloud go to the same route, or is it a little tougher for them?

Du'an Lightfoot: 12:25

I think it's two paths. No matter what cloud provider you go to, there's some type of foundational certification that's for AWS going to be the cloud practitioner. Cloud practitioner is going to be teaching those building blocks. It just give you a quick, high level overview for, let's say, the stakeholders before someone that just want to understand what the cloud is, whether the service is, whether the building blocks are the cloud. But then what should evolve beyond that? And you get that certification is going to be the solution architect which tells you, okay, here are the building blocks, but here's how you build with these building blocks, and it kind of tells you how to architect those solutions. And so now you not only understand what the building blocks are, but you have some understanding of how to actually to put those building blocks together to architect solutions for customers.

Du'an Lightfoot: 13:12

Now it's a fundamental thing. If you study just for the certification, then you understand it from a book level knowledge. You actually have to go in and build something to really get that applicable knowledge that you can say, okay, I understand. If I create a VPC, I understand the route tables and everything that I need Now understand. Okay, if I want to create some type of application that's highly availability. I can use some type of load balancer what is ELB or ALB or something to provide that connectivity outbounds from the internet. You understand all of these different built building blocks and how to secure, at a say, an associate's level, which would be like the CCNA mentioned.

Alex Perkins: 13:52

So I got basically two more observations and questions here.

Alex Perkins: 13:55

But the first is when you were just going through your list of all the things, you just condensed like 15 years of knowledge into like a couple sentences, right. But it also, I think that speaks to like how the cloud has changed things, because you're talking about like you have to learn the compute right, like you can look at the cloud and you can learn all the different compute types, you can learn the different networking types, you can learn the different storage types, right. So I think it's almost like a completely different field, almost right, because on-prem everybody came up with their silos. It was like, oh, there's a server admin, there's a network admin, there's a storage admin, right. And now it's like when you come into the cloud, everything is so easy and consumable, generally speaking, right, but it's like that's the starting point is, you can just go into all these abstractions right away, instead of like having three siloed tracks and having to work your way up to an expert level basically in each one. So it's just more of an observation than a question.

Du'an Lightfoot: 14:55

Yeah, a lot of the heavy lifting is abstracted right. I remember my last on-prem networking job. We were standing up paloes in between our data centers and so those had to be physically placed right. So that takes time. And then when you're doing an upgrade, of course you got to manage the routing. You got to flip inbound and outbound between the paloes and the ASAs. You have to do all of these things while ensuring that you don't lose connectivity, which in the cloud it's kind of like the same thing, but the physical part you don't have to worry about.

Du'an Lightfoot: 15:32

And then the other part is the infrastructure is cold. Like you can actually do this, not just from infrastructure is cold standpoint, but you can actually stand up in an entire new environment. Like that. You know what I mean. You don't necessarily have to fail something over. You can just create a whole new environment and stand that up and fell over to that, Just change your DNS or change your routing, depending on how you want to do it, and your new environment is up. So it's a different approach and I think just understanding the approaches that are out there with this canary deployment, blue, green, brownfield, greenfield and some of those terms on how you want to approach your infrastructure, but the infrastructure itself, I mean, it's kind of a firmware, it's one of those things that is there for as long as you want it and then, once you don't want it anymore, stand up and do it.

Tim McConnaughy: 16:24

Yeah, it's timely. I was on a design call not too long ago with someone another architect, and they were showing a very complicated. This was actually. This was not NABUS, this is an Azure, but it's relevant. It was a bunch of like UDRs pointing between different I think of like firewall sandwich, like we would do on prem, but like in the cloud, right? So you got subnets, that with route tables, udrs pointing at interfaces for firewalls that come out the other end and do another subnet.

Tim McConnaughy: 16:57

You know right, really really complicated the way the customer built this and the question was how are we going to migrate this customer off of it? And I was like here's what you're going to do. This is really hard. You're going to build it differently, you're going to build the whole thing up again differently and you're going to put you're going to migrate them to that. You're not going to make this work. And he was like, but well, that's dumb. You know what, if we have to do that, why don't we? Just? That could always be our answer. Maybe we don't ever migrate anybody to anything. And I'm like welcome to the cloud. Like well, that's like how we. You know we could do that Not all the time, right, but I mean you don't have to try to figure out how to do these complex migration patterns that you had to do on prem when you're actually pulling cables and moving and failing over fiber and stuff. You can build like you just said. You can build it better and then just move your workloads over to it.

Du'an Lightfoot: 17:45

Right, it's all about what the customer needs, right, whatever your business needs, and understanding the solutions out there. And sometimes you may not know, but you make and work with a solutions architect at AWS that could help you figure out. Okay, do we lift and shift or do we do some type of something brand new in the cloud? But one thing you shouldn't do is try to build the same thing that you have on prem in the cloud, because it's probably going to be expensive for one.

Du'an Lightfoot: 18:09

It's very if a tool is not always the most efficient way to do it, which which brings me to another point, and I don't think I mentioned this.

Du'an Lightfoot: 18:15

As far as, like, the skills you need to understand, I think part of that is going to be cost right. One of the things that we didn't have, one of the things we did have to do on prem, is understanding. I would say, like when optimization and things like that capacity planning we had to do these on prem and I would say, in a cloud it's not necessarily capacity planner, but you do have to understand costs. I think to be like a really valuable engineer, I understand it. Okay, if we have not gateways and one availability zone, but we have two availability zones and we're transferring, we're moving data across one availability zone to that gateway, there's going to be a charge there. So understand that all those cost points in the cloud can be something that you can help your organization save. Save like purchase dollars and save money in the cloud and do some cost optimization at that standpoint. So again, it's a new way of thinking on how we approach developing and building our networks in the cloud.

Chris Miles: 19:16

Yeah, I think that's a good point. I think I feel that's been one of the most difficult things for a lot of standard engineers even myself included with migrating to the cloud is being more kind of cost conscious with relating to throughput and things like that, because half the time, whenever you know someone's doing a migration project, they'll be like okay, well, you know, we got to consider this. You know, how much are you sending traffic wise today? The answer is always nobody knows. So it's always like just like a ballpark estimate that could be completely off. But you know, it's like you said about the visibility piece is like how are you going to monitor these things? It really, it really factors into the, the overall cost of the solution.

Du'an Lightfoot: 19:58

I can share a talk for a real event that kind of goes over data transfer charges and, if you would want to put it in the show notes, I think it'd be helpful for the audience to go through that Because, again, that's a new way of thinking and that's something that is, I think, be helpful to the community.

Tim McConnaughy: 20:15

Yeah, we've done, actually done, a one or two episodes about cost specifically because we completely agree, right, that, no, no, no, I want to get those in the show notes, but I mean, is that we've, we've? We agree with you, right, like? We believe that cost, engineering for cost is definitely something that's a skill that is very different on prem, right you different, like you said. Like is capacity planning and stuff like that on prem. Here it's very, very different. Engineering for cost is important.

Alex Perkins: 20:40

Well, and I mean, wasn't Werner's talk about the frugal architect right, like that's the whole right? His whole thing from Werner Vogels keynote was all about was a lot of it not all of it, but a lot of it was about being frugal and thinking of things like, like cost when you're, when you're architecting these solutions in the cloud.

Du'an Lightfoot: 21:01

I love Ernest talks. Every year he comes with it. He lost a weight to. He looks good.

Tim McConnaughy: 21:06

Something we, something we've been talking about a little bit, but I feel like we really haven't like hit it dead on yet is when we talk about the skills gap between, say, network engineering in the cloud.

Tim McConnaughy: 21:17

But really I guess that it's really any technology If you're, if you're a network engineer right now and you know you got to pick up cloud and that's going to be part of your thing, and I feel like the cloud and all the clouds, I think all the CSPs have done kind of a disservice on the training side for a lot of the network stuff. And what I mean by that is, if I didn't know networking at all, I don't think I could learn networking from the training that I get from the CSP. If that makes sense, Right, it barely touches subnetting. That's my opinion and that's all the CSP training that I've gone through.

Tim McConnaughy: 21:57

I, having said that, I kind of to circle back a little bit to what Chris had said earlier. Like you know, we learned probably on what's probably the industry, honestly, the gold standard of training for networking, which was CCNA, which is, I think, maybe still is. I do feel like if you started from zero today and you decided that you wanted to do cloud and you were trying to learn networking from zero, I still I do still feel like the CSP training would not be sufficient. It might be a sufficient for cloud. I don't think it would be sufficient for, like, anything beyond the kind of basic networking in the cloud. I don't know. What do you, what do you think?

Du'an Lightfoot: 22:37

It's tough because, like when you look at the reinvent talks and for networking, we started at level 200, right, so it's already assumed that you have some basic understanding of networking. Yeah, also, if you're talking about fundamental networking, when I start we're starting basically at layer three. Layer one, layer two are kind of removed, so we're already at layer three and then much of that is still kind of abstracted in how we do art, but not really you know what I mean. So it's kind of different in the cloud. So explaining it from a traditional networking standpoint would make sense, but I think your better serve by learning that is someone special somewhere that specializes in that and then coming to AWS and then picking up from the foundational tools that you already have.

Tim McConnaughy: 23:27

Yeah, that's fair. And I realized after I asked the question that sounded like like maybe I was talking about, you know, like any training that you or AWS have done. Really I was speaking, I was in my head of what I was thinking was specifically like studying for the certifications, like the certification training the stuff you see in the certs is very, very basic, is kind of what I was meant. The actual ADB is docs and the stuff like the you and the other solutions architects and stuff put out is amazing, like it's really really good, right. I was thinking specifically of like when I took, well, when I took the CCP, you know, a couple of years ago, and then when I took the solutions architect, it wasn't until I got to ANS. To be fair, ans is like a specialty cert, right.

Tim McConnaughy: 24:11

And that's a serious cert, right, like you know, that's a real, that's a real deal. So, yeah, anyway, just from a skill gap perspective, what I mean is if I'm a hybrid engineer or a network engineer and I want to become a hybrid network engineer meaning hybrid cloud or both, or I want to do both I do feel like I would be empowered. In fact, I was right, like we did it all three, all four of us did it to make that jump right. But there's a lot of people out there still today who were, you know, they look at it and they look at it like it's a mountain that they can't climb, you know. So I'm still trying to, like I've and you know you were both content creators, right so I've been scratching my head as well like how do I make this accessible? Like, make it like actually this isn't hard, just like. Come with me and take the few steps. I don't know. I hope that makes sense.

Du'an Lightfoot: 25:05

No, you're making sense. I've had a lot of conversations with this and normally when I whiteboard something or diagram something, people get it. But I think the challenge is, if you don't use it, you lose it like, oh, it makes sense, but then you go back into your organization, you're just standing up a VPN to 169.254, you know what I mean or something like that to AWS, rather than actually seeing what's on the other side of that tunnel.

Du'an Lightfoot: 25:29

So it's one of those things that if you're not getting hands on, you know it can be kind of hard to apply. Rather than traditional networking, you may get access to the layer two in your infrastructure. You might be a land engineer or something like that where, okay, someone can take me under my wing, I can actually see what they're doing. I can go rent the switches and routers, or I can get the switches and routers off eBay or Craigslist, go home lab it out for myself. But now, with the cloud, I got to kind of log in. Somebody told me to worry about costs and I got to think about these things. So you don't know what. You don't know. You may have some fear around that. So I think really helping people understand okay, let's put in some budgets. Let's understand that. Okay, if you deploy something in the cloud, let's tear it down. All right, now here's some guides. Here's some fundamental things you need to know about Bit on the VPC, we have a resource map, resource map that'll show you. Okay, if you're in this region, you click on this VPC, here are the resources, here's your subnets, here's your route tables, here's your internet gateway, your S3 endpoints, your NAT gateway. Everything is right here, clearly, so you can see it right and then understanding that, okay, if you go to a different region, the VPCs are going to be totally different. And just getting them comfortable going into the console and once you get them comfortable there, building something.

Du'an Lightfoot: 26:44

Okay, let's pin up an EC2 instance, let's launch that and do a basic Apache web server so we can just see some web text. Oh, I deployed something in the cloud, this is awesome. And kind of just build on that. And now let's stand up a site to site VPN to my house. Maybe you have a ubiquity router or maybe you want to decline VPN to your laptop. Let's set that up and just kind of build on things that they can actually do. And then, once we're done, let's tear it down and then look at the charges that we may assume from running these services so we can understand okay, I got idea. Next time when I go through this, I'm going to look at the pricing calculator to really understand what I'm going to be charged for this lab that I'm going to be deploying and not have to understand. Understand how this all works, and I can really think about when I'm in an organization, the things that I should be thinking about when I'm building. That's my thoughts, but I'll let you hear your thoughts.

Alex Perkins: 27:36

This is absolutely great call out about starting with, you know, learning some of the cost stuff, because that's something I think I hear a lot I'm sure you hear it a lot from people that are just kind of dipping their toes into cloud is like oh well, I'm worried about how much is going to cost. You know, I hear all these stories. There's that meme that's always going around about like, why are you homeless? Right, because I left an EC2 instance on right. But it's great that you point out that that's that should be included, like as soon as you start diving in.

Alex Perkins: 28:05

You're talking about, like the billing calculator, the resource map, like all that stuff is super valuable and I think that's not something that I hear many people talk about. When they're talking about that, everybody usually is just like this will be a couple bucks a month. Just go ahead and do this. But nobody understands. Like you're calling out why that's even a couple bucks a month, right, like you should be looking at those things and creating billing alerts and all that other stuff. So great, great call out.

Du'an Lightfoot: 28:30

I've talked to so many people directors, people that run organizations in the cloud and like, the number one thing that they struggle with is understanding, like, not gateway charges, because that's going to be on the EC2 rather than on the networking. So just understanding some basic like that, right, that if you're running in that gateway, it's going to be on the EC2. Or if you happen to deploy or create a domain under Route 53, every domain you create is going to be a 50 cent charge a month. You know what I mean. So, understanding that the different charges that you see on these bills although they may be 50 cent, maybe a dollar or two, but no one why they're being charged for that and how to find them and being able to read your bill when it comes in every month, I think.

Du'an Lightfoot: 29:17

Starting there and then understanding, okay, I'm doing this on my own, but there are platforms out there that will give me access to the cloud environment, right, skill bitter on AWS, we have cloud Quest. We have all these different tools out there that you can go through, you can build and you don't have to worry about using your own account, right, I think? A Cloud Guru or these other sites as well. You can use those sites, pay the $29 a month or however much it is, you can access the cloud environments. You don't necessarily have to worry about coming out of pocket. If you're worried about costs, there's ways around that.

Tim McConnaughy: 29:49

Well, and those great call out on the like A Cloud, Guru and those type of training sites. Not only do you get to use their environment, but they build a lab to teach you something right, Like you don't have to go model through it yourself. Right, the lab is right there with the guide and all that stuff.

Alex Perkins: 30:05

Yeah, right, if it's like something complicated. Right, it can spin up pieces that are already in place that you might not have learned about yet. Right, just so that you can play with something particular and you don't have to worry about building out the entire architecture.

Tim McConnaughy: 30:18

Skill builder too. I mean it's like you said, like I haven't tried Cloud Quest. Now I keep hearing about that from a few of my friends. I got a. I mean I just for, I'll just check it. I just want to check it out.

Du'an Lightfoot: 30:28

It's actually pretty cool. You go into like this simulated environment and you have these different quests where you have to architect solutions to say, hey, I got this problem and we need to fix something we can't figure out. This is what we have now and you, as a solution architect, it gives you the type of questions that you should be asking to formulate the solution. So it's really it's kind of nerdy, but it's still kind of cool on how it gives you like an idea of how you should be thinking about architecting solutions in the cloud. I think that's really awesome, yeah it makes?

Chris Miles: 30:59

it makes it like a game, Kind of like how Duolingo does it right. It's just teaching you a different concept and gamify the learning Exactly.

Du'an Lightfoot: 31:08

And there's one for, like, I think, every domain. So there's it. It's like the cloud practitioner, cloud Quest, there's the networking, there's the data analytics, there's, like all these different quests, security.

Tim McConnaughy: 31:20

Awesome. We'll have to get that one in the show notes too, and I'll have to spend some time playing with that. It sounds fun to me. I'm just one of those guys a nerdy guys, I guess but it sounds fun to me.

Chris Miles: 31:30

I can verify.

Du'an Lightfoot: 31:31

My tition is free so I'll get you that Nerd so how about for that?

Tim McConnaughy: 31:36

If it wasn't cleared by all the stuff behind me.

Chris Miles: 31:41

So I want to circle back a little bit to the skills gap thing, and we're talking about this is probably going to be speaking to like a niche of a niche, whereas like someone getting into the industry now that wants to be a networker and also wants to work in the cloud probably not a ton of those people out there in the industry right now.

Chris Miles: 31:57

But let's say, if you were, you know a lot of these like networking focused certifications, at least from the, you know, from the different cloud providers. You know they all do kind of have this understanding that you come in with a level of networking knowledge that you're expected to obtain elsewhere, right. And if you're new, if you're convinced that you want to work directly in the cloud, if you're going through a CCNA course, there's probably so much bloat in there that is that you don't need right for that particular environment, because, you know, one of the best things I love about the cloud is there's no layer two, right. And if I have to sit there and learn layer two for hours on end, for weeks on end, I'm probably kind of, you know, be bored to death. So that's like, how do we feel that newer people to the industry should tackle that? So they kind of like build a better mind map of how networking works in the cloud and what constructs they need to learn and tackle them individually, or how do we think they should approach that?

Du'an Lightfoot: 33:00

A better mind map. I do like that approach Because I was actually talking to a community builder at ReInvent and he's architecting some amazing solutions but he ran into a wall where he had to architect to some branch offices that were running SDWay and he was just like I just don't understand his own pram stuff. But he really understands cloud networking. So I think you kind of have to understand or focus on where you are right. So, if you want to go to the cloud, understand the cloud and then, once you understand the building blocks and understand the hybrid connectivity, the VPC, the VPC connectivity and those things in what? Hybrid connectivity being? How to connect to the internet, how to connect to other regions, how to connect like a standard site to site, vpn not necessarily direct connected or anything like that, but just understand it. How to connect that VPC to connectivity, how to connect to the internet and from the internet. Understand that first and then, as you are faced with new challenges, I would say on-prem and your knowledge gets a traditional understanding of the transit gateway right. The VGW understanding of these tools. Direct connect may be a challenge because getting a hands on that may be difficult, but at least read about it right. Understand Route 53, because we do a lot of routing with Route 53. So, understanding that as well. So, once you understand, like those four pieces, then start exploring on-prem, because it's kind of hard to just okay, let's throw all the dots on the board and kind of figure this out.

Du'an Lightfoot: 34:37

I don't really have access to SDWAM, but I want to learn about it. Well, you can get access to the cloud. You can connect VPCs. You can connect to other regions. You can you know, you can peer transit gateways. You can do VPC period. You can do all of these things in the cloud. Now, right? So start there and then, once you build that foundation and you're, let's say, your task was something in your job, then scale out to the community and ask for advice on how to you can learn things faster, because I think having the community as you learn does help you to be able to skip some steps. As far as learning the hard things, which somebody that's been doing it for years can explain it simple, right? So I think I would start in the cloud and then work my way out. That's my thoughts.

Tim McConnaughy: 35:21

Yeah, I think that's fair. I think that's really good advice actually, especially if someone is starting now and they want to start in the cloud, like that's. Yeah, because it will take. Also, it'll take a lot longer to pick up the extra cloud info, like the knowledge of SDWAM and on-prem and all that, than it would take to pick up the cloud side of the house. You know, by design right, it's made to be more consumable and easier to understand.

Chris Miles: 35:43

Kind of the same thing like if you're getting into IT for the first time, you got to get in just work at the help desk, right. You're doing shit that you don't really probably enjoy doing, but you find what you want to do and you expand on that, you learn on it, right. So I think that is good advice. Just at least get your foot in the door, or at least you know, get some knowledge under your belt and then work your way out, it started.

Alex Perkins: 36:03

Well, yeah, like you said, Tim, get started. That's the easy, like it's as simple as that. Right, it's so easy in the cloud. We've talked about this before. I think like one of our first episodes we were like CDN right, like you can just have a global network in no time in the cloud. So it's like yeah, if you're worried about being able to lab it and get hands on Dwan, you're exactly right. Like just go to the cloud. You can spin up literally anything that you want to study very quickly and very easily.

Tim McConnaughy: 36:31

Except for direct connect yeah.

Du'an Lightfoot: 36:36

One of the things I want to like everyone to think about is that there are companies that are cloud native, just like there are companies that are Slowly on print. There's cloud. There are companies that are cloud native. So if you want to focus on a cloud and you become really good at that, you know you could actually do that and work for a company where you're completely Cloud cloud native.

Du'an Lightfoot: 36:55

I know people that started their career slowly in the cloud and never had to do on print for the first three, four years of their career, because the application is living the cloud.

Tim McConnaughy: 37:05

Yeah, yeah, absolutely I want to.

Tim McConnaughy: 37:09

I know we're starting to run out of time and I really want to get to this last piece and I know you, you do as well, duane, so let's.

Tim McConnaughy: 37:16

So, one thing I wanted to and this goes right with the skill gap thing, right, which and and on the back of reinvent as well, which was like the AI extravaganza, something we've been talking about I don't know if we've really gotten to talk about on the podcast very much yet, but that I personally have been talking about in the community and stuff is how, you know, one of the things we need to do with AI and this is my opinion, I'm see what you think about it is To make sure to safeguard either safeguard or just figure out how we're gonna deal with the fact that you know, as more and more things are Handed off to, like entry-level type of work or operations level work is handed off to, you know, chat bots and AI and whatnot. You know, how are we, how are the people that are junior today, that want to be senior tomorrow, going to be picking up that Expertise, like learning those ropes and making those mistakes and whatnot?

Du'an Lightfoot: 38:08

Are we cutting the bottom rungs off of the career ladder, you know, to save money essentially, I don't know what do you think man is moving so fast that if I give your answer today, tomorrow it's gonna be totally different, like?

Alex Perkins: 38:21

Time to say ours.

Du'an Lightfoot: 38:22

I'm gonna be 1000% honest. Every time I think about generative AI, it's a new, something new that just makes me like, yeah, what I was saying last month. I had no idea what I was talking about because it's something totally different. But what I will say is that that there's a couple pieces I think we should use it to learn. I think we should use it as an assistant to help us learn faster, to simplify things that are complicated, break it down for us, and then we should also challenge it to give us the reference of the information, right. So if it's explaining something and you're not sure if it's right, ask it again Is this right? Right, because then it'll go back and set and verify whatever just told you and say, oh no, I was so wrong. Sorry about that. Let me give you the correct answer. You know you may have to reiterate on that a couple times, but I think you should use it for that. Whether it's learn a python, whether it's learning networking, I think you should use it to learn and understand how to ask prompts, understand it something that is basic to that. The second thing is that, like our documentation, the documentation that organizations is something that the ball gets drops often right, like Documentation I have rarely seen done well and done accurately, consistently. And so if we can leverage general AI to even look at our Ticking the system, to look at our knowledge basis and kind of curate all this data into a database that we can leverage To help those entry-level technicians get the answers they need, what, a, what they need it right. A ticket gets open. It says okay, there's a when link that's flapping. Well, I've seen this ticket about four different times. Here's the most lucky reason. You know I've already contacted the service provider, etc. Etc Now send that to tier one so they can analyze it, triage it, see if any steps are missed and then communicate with the stakeholders or just have general AI do it right, because you can put those safeguards in place. Like there's part. There's ways to put these safeguards in place. So I think that's one piece. I think we should use it, you know, for learning and to analyze data.

Du'an Lightfoot: 40:25

I don't know if you ever had a hundred thousand lines of locks. It had to filter through those laws to figure out the root cause or some type of root cause analysis. Oh yeah, that is painful, but have you ever took generative AI and had it? Look at some locks If I had time in. If y'all want to look offline, we have this tool called party rock, and the first thing I built in party rock was a Cisco configuration Configuration config configuration analyzer and what it does is you put a Cisco fit configuration in it and it tells you everything about that configuration. You can carry it to find the source destination for routes. You can see if no matter if it's in scope or out of scope, but it tell you that access control is any vulnerabilities and the configs as far as insecure passwords. It'll do all that same thing with ACLs. If you're filtering the laws of ACLs. It's great at that.

Alex Perkins: 41:14

I was just gonna say ACLs man I. That would be so, so helpful. Or like a wire shark analyzer oh my goodness.

Du'an Lightfoot: 41:24

Easy, my bad, go ahead. I'll get excited, I'll be quiet.

Tim McConnaughy: 41:26

No, no, no, no, you're. You know I'll let you speak more. I wanted to mostly agree with the idea, with the what you're saying, right, like AI would be any kind of large language model. Anything that can do a analysis would be really good at taking data and making Sense of it and presenting that as like hey and an insight. Right, you're taking actual insights, or where's that that mean? Where you go from, like you know Points on the, you got points on the paper and then use like here's the data and like here's the insight, and you see him draw On pictures like with the dots and here's the actual insights from it.

Du'an Lightfoot: 42:02

That's like one piece right lose using a large language model. But you can actually take that large Vegas model that is like general Understands how to do. Look at it as like a Swiss army knife that could do a whole bunch of things. But you can actually fine tune that for the data in your organization. So it understands your documentation, understands your network design. It understands all these things. It can answer and analyze routes. Analyze access. Control is Really well because it actually understands your infrastructure like this is being done. This isn't something that's like this is actually being done. To me that's that's like exciting, because it can make a lot of the challenges that we face in that working simpler.

Tim McConnaughy: 42:43

Yeah, so the concern I Agreed with everything you just said. The concern I have and it's really goes back to the original question or the original Thesis statement I had there was you know how long will it be before a business decides that that is good enough, as, like, say, a help desk, right, and so somebody who was looking to get an entry-level job on the help desk, to get their career started and like, get that experience, just, that position doesn't need to exist anymore, right, we automated it away, essentially, and that I'm not saying whether it's good, bad or ugly, you know, that's, that's technology. Right, at the end of the day, that's technology and that's just how it is. I'm wondering okay, so now that person and and I'll let I know you what I know but so now that person like, doesn't you know that there's no entry, I guess how long before there's no entry-level right, and then what? How do you get the experience to become Senior at that point?

Du'an Lightfoot: 43:36

I guess is where I'm thinking that is a really tough, a tough answer. But I read a lot of the white papers and I'm gonna send this all to you now, okay. But Open AI has a white paper that's called GPT's. Are GPT's an early look at labor impact on potential a, a potential of large language models and basically what they look? That is how large language models is affecting the entire industry. Basically, these large language models generate high is affecting 90% of all jobs. The only job that is not affecting is things that you do with your hands that makes sense.

Du'an Lightfoot: 44:12

And then they say, abide, the year 2029, it's gonna be just as good as, like, a mid-level developer. You know what I mean. So I think I don't think it's necessarily okay, it's gonna be. I'm taking the interlevel jobs and those are just gone. I just think that it'll be different on what those jobs actually look like.

Tim McConnaughy: 44:33

That's, that's a good answer. Yeah, no, that's, that's interesting. Anyway, it's just, it's, it's, I'm, I'm, I'm not bullish or or bearish on AI, I think. I think technology is technology and people it's coming, whether you want it or you know, whether you like it or not, and and whatnot. I am curious about some of the Considerations that I am a little bit concerned that you know, all the corporations are kind of like not paying. It's not Everybody's in a race, right, and so the race tends to run over everybody. It's on the track, if you will, I.

Du'an Lightfoot: 45:06

Think, yeah, I feel like we've seen this before, though, right with like the dot-com boom or something. Yeah, everybody just had to get a website. I think we're like the infancy of something that's new. What the end result looks like, I have no clue, but I mean it's here and I don't see it going anywhere anytime soon.

Chris Miles: 45:25

Yeah, I think it's it's hard to predict, just like you said to on it's like it's it's so fresh and and there's, you know, a million theories we can throw out. And I think, tim, to kind of go back to your, your hesitancy about it, I think it is probably going to Make infrastructure workers like a more IT infrastructure professionals, probably more of a niche thing than it is right now. It's not going to be as widespread like that. I think probably set some point. Getting into tech is going to be being a developer, right, because all this stuff is either abstracted for from you or it is automated for you. Even the troubleshooting of it is automated.

Chris Miles: 46:05

You know it's like, yeah, I mean you see Amazon Q, right, and it's just got launched. It's like how do I, how do I build this? Like that it it dumps it down, not dumps it down. I hate to say it like that, but it makes it a very consumable and bite-sized thing for you to build Networking, security, things like that, right. So I feel like it's it's just gonna, like you said, to want to change the way these jobs are done. There's still gonna be people that are gonna need to be very fine-tuned. You know experts at that craft and Building it for the consumption of other people. But you know, I feel like we just probably got a hopefully, the hopefully this boat. This boat stays afloat until we're all retired and then and we'll be alright. But I think it is gonna change quite drastically.

Du'an Lightfoot: 46:47

I Think you you hit the nail on the head, chris, when you said expert. I think the days of being kind of like a journalist and, you know, having a long, lasting career as Someone that is comfortable just having a job. I think tech is going to push boundaries to where, if you want to like growing your career, if you want to get like let's say even I hate to speak money, but just to say get to a certain dollar amounts, that You're going to have to be an expert, to where, if the model does give some irractory information or there's something incorrect going on with the data, that you can analyze this and see it, and those will be the people that will stand out and be the most valuable. I think that's if I am to give some sort of opinion.

Chris Miles: 47:32

I was just going to say, yeah, I think, to your point, duana. You got to be able to, you know, be if there will be a fewer experts in the world that there is right now. But you also got to be able to adapt right, because, I mean, the job market's got to be Catering to what your, what your skills are in that expertise is right now, because, like I'm sure, we have, you know, several Cobalt developers out there right now that aren't that, aren't getting work for that. So it's like, you know, there it's you got. You got to be able to switch with the times right.

Alex Perkins: 48:01

Yeah, I, I just I want to point out I also think that you know this is like we're in the middle of this. I like humanity has a history of kind of like new things coming in and disrupting things and Finding ways to adapt and grow through it, and it's this is just the one that we're going through. It's kind of like my point here is like we feel like this is going to change everything, but who's to say, in like 30 years, people look back and it's it's like something that we look back on 30 years ago. Right, like it's it's hard to say and maybe we all just adapt through it and things do change.

Du'an Lightfoot: 48:34

Like like the one saying One thing that I did I don't know if I've made it clear, but when I said, like the dot com boom is, one of the thing about the dot com boom is that, um, it may Dot coms accessible for everybody, but only the few really knew the opportunity that was there. And I think genovii Makes building accessible for everybody. Right, we're going to have new fortune 5100, 100 companies that may have 10 people, you know, because you don't need as many people to build solutions. Like it's a new approach. And so for those that want to build, this opportunity for you to build your own industry or whatever you want to build, without having all of the pieces that you needed previously, because now you have genovii to kind of help Do the things you know today that you would say, uh, finally, I had a lawyer to review this document, or if I had somebody to focus on my documentation or to guide me through this code or to do all of these things now, like it's kind of accessible for anybody to build a team of assistants To establish whatever dream they had.

Du'an Lightfoot: 49:45

That's my thoughts. I don't know.

Alex Perkins: 49:49

Man, you're this, you could be like a profit or this could fall flat. But like your quote about like there might be fortune 100 companies with 10 people, that's just like that. That's a interesting way of looking at it and just makes me really really think. I just really like that Because it could be true. You never know.

Tim McConnaughy: 50:07

Yeah, we're scratching the service, regardless of how it goes. We're definitely scratching the service at this point and I don't think any of us knows where it's going to look like in a month much less you know, 20 years from now or even sam altman will be back fired again, right, that's right.

Tim McConnaughy: 50:20

Yeah, what is it? Tuesday? Or? Yeah, sam altman's fired again today. No, I'm just I'm kidding, no, I, I mean that's. That's the kind of weird crap that keeps going on, though so, yeah, I'm very interested to see. I am, I am a little cautious, but I am the tech's here and we're all gonna have to figure out how to deal with it. We're gonna have to figure out how to make use of it, how to adapt to it. It is what it is. So, anyway, I know we're out of time here. Let's go ahead and wrap up. Um duan. Thank you for being on. Any closing thoughts you'd like to?

Du'an Lightfoot: 50:50

yeah, um, tim chris, alex, Thank you for having me on. I wish you all much to say so and cable to the cloud for me, for the community. I think this is a like the best time to be in tech, because information is still freely available, the resources are widely available. There's people like tim chris and alex that are helping, doing whatever they can to help you all grow and advance in your careers. I think if you, whatever you want to do, take advantage of it, find the right communities, the right resources, ask questions, learn, be curious and just don't stop, go after it chris, alex, you want to anything for you guys, chris, you want to kick us off.

Tim McConnaughy: 51:27

So you see that you're going to talk over alex anyway.

Chris Miles: 51:30

Yeah, all good. Um, no, I think this has been, this has been absolutely wonderful. So, yeah, thanks to one for coming on. Um, we, we've, we've had you on our list for a while as someone to have on the pod, so I'm glad we, we finally got to make that happen. And, uh, I think you, I think you really Touched on something in that last closing thought that you had about a lawyer, uh, reviewing documents and how it relates to AI, and I'll just spoil it for everyone lawyerai is already taken, so you can't have that.

Tim McConnaughy: 51:55

But let's go out.

Alex Perkins: 52:00

Thanks, uh, no, I, I don't really have anything to on. This was. This was awesome. That was a hell of a sendoff, uh, so I just wish you best of luck, man. It's been. It's been great watching your, your journey and and just following what you do and the the content you produce. So thank you.

Tim McConnaughy: 52:15

Thanks for coming on and, and I mean hopefully, we can have you on in the again in the future. I do love what you said about this is the good. This is probably the best time to be in tech. I know we just painted this doom and gloom scenario where, like, everybody's job is gone tomorrow. But he's absolutely right. I mean, it has never been easier to learn anything you want to learn. Right, when we were up and coming, it was tough, right, and especially before the before the internet was damn near impossible, I swear. But, uh, now, I mean right now, yeah, and it's never been easier. That's absolutely agree with you. So, um, yeah, so this has been, uh, cables to clouds. Thanks for joining us. Obviously, I'm your host. Him, uh, chris alex douan. Thanks for being with us.

Tim McConnaughy: 52:56

Um, if you liked what you heard tonight or saw, depending if you're watching on youtube, uh, please feel free to subscribe to our youtube channel or listen to us in your favorite podcatcher and we're out. Hi everyone, it's tim and this has been the cables to clouds podcast. Thanks for tuning in today. If you enjoyed our show, please subscribe to us in your favorite podcast catcher, as well as subscribe and turn on notifications For our youtube channel to be notified of all our new episodes. Follow us on socials at cables to clouds. You can also visit our website for all the show notes at cables to clouds dot com. Thanks again for listening and see you next time.