Episode 059: How to Decide Between Azure and AWS?

Mike Pfeiffer on January, 21, 2020

In this episode I catch up with fellow cloud architects Nicolas Blank, Warren du Toit, and Chris Goosen to discuss the determining factors of choosing Azure or AWS.

With the explosive growth of cloud services, many organizations suddenly find themselves having to manage multiple cloud environments, either intentionally or by coincidence due to mergers, acquisitions or other such events. With Azure and AWS being two of the most popular choices, we talk to Azure MVP Mike Pfeiffer who is also AWS certified about the parallels between the two services, things to consider when running two clouds and discuss if multi-cloud makes more sense as a transitional state or long-term mode of operation.

Make sure you check out The Cloud Architects podcast from Nicolas, Warren, and Chris for more episodes about cloud, technology and the people using it.

Full Transcript:

Warren du Toit:
Hello everybody and welcome to Microsoft Ignite 2019. We are in the pod cost booth area booth number four. We have four people in the room. I’m Warren du Toit and I’m with my co-hosts.

Warren du Toit:
And today we have a very special guest and we have a contentious subject. We say the words A, well, sorry, it’s an acronym. The words Amazon Web Services and we’re here with Mike Pfeiffer. Mike, what’s happening?

Mike Pfeiffer:
What’s up you guys, I’m super pumped to be here. Really appreciate you guys having me on and hope you guys are having a great conference so far.

Warren du Toit:
Oh yeah.

Mike Pfeiffer:
It’s going to be awesome.

Warren du Toit:
Today has been crazy.

Chris Goosen:
My feet are already sore and it’s like day one.

Mike Pfeiffer:
Well it is day one. Yeah. You’re going to need to take next week off probably to recover.

Nicolas Blank:
I wish.

Chris Goosen:
It’s the truth right.

Mike Pfeiffer:
I was waiting for you guys to laugh at that.

Chris Goosen:
But it’s actually true and you know we’re back together again too, right? It’s been a good, I don’t know, like six or seven months since we were actually like in the same room. No, actually we did record something at summit, so we were kind of-

Warren du Toit:
Oh yes, you’re right.

Chris Goosen:
It’s always good when we’re back together. But we’re really pleased to have you on the show and let’s get some introduction rolling here. So you have a particularly interesting, I guess history and background in tech and what you’ve been doing over the last few years and we’ve known each other for a while. We were just talking, we hadn’t seen each other in probably six or seven years, so time goes by so quickly. But give the folks at home a rundown of who you are and what you do.

Mike Pfeiffer:
Sure. Yeah. So I started 20 years ago, started in the late 90’s and I was basically just a kid living in Albuquerque, New Mexico. I was not a computer science major, actually dropped out of college a couple of times. And when I was living in Albuquerque I got lucky and got a help desk job that I was not qualified for. My first phone job ever was America online. Remember the old school AOL dial up. So that was kind of the early kind of entry point into tech.

Warren du Toit:
That’s when they designed the web interface for AWS.

Mike Pfeiffer:
Like the backbone, right? Of everything we’ve ever done. But that was a cool kind of experience for me to break into an emerging industry. The internet was starting to take off and Microsoft was starting to take over client server networking. At the time NetWare owned kind of like the server side, right? And it was a different world, but I was lucky enough to get in. I got obsessed with technology at that point and I’ve always kind of had to elbow my way in because at least that’s what I felt in that era, right, because I wasn’t a college graduate. I wasn’t a computer science expert, but I was passionate and I think for me like that’s been the reoccurring theme in my career. I’m always trying to learn the next thing. And I think that that’s, if you look at my past, that’s kind of what I’ve done.

Mike Pfeiffer:
I’ve never worked anywhere longer than three and a half years. After the first couple of years of doing help desk, I went into IT consulting because certifications propelled me into being an MCSC and being able to play that game. And then I spent a couple of years as a dot net web developer, which was interesting. And I quickly learned that writing code is fun, but it’s not something I want to do 40 hours a week. So I went back to infrastructure and I’ve been kind of playing both ever since. And you guys know me well from the Exchange world and so when I was doing-

Nicolas Blank:
That’s how we do know you from the Exchange world as the Exchange guy. So knowing you as an Exchange guy, I’m surprised that you have this coding background.

Mike Pfeiffer:
Yeah, I think that for me it’s just my personal interest in it, I know it’s not for everybody and that’s cool. You know what I mean? But I was always fascinated with web development. Even in the late 90’s I tried to build websites when I was doing the help desk stuff and I always just did that on the side and it ended up being something that I did for two years professionally, excuse me. And then after that I was like, all right, I don’t want to do this 40 hours a week. I just want it to kind of be a side hustle and that’s what I did. And so when I got back into infrastructure after doing a couple of years of dot net web development in the very early days of dot net, like right after it came out, I noticed that every customer that I was working with had Exchange servers. I was like, “Huh, that’s an interesting angle for me.”

Mike Pfeiffer:
And so I pushed the chips into the table and I was like, “This is what I’m going to focus on.” But as you guys know, that takes you into AD, lots of infrastructure, lots of architecture, and really everything, right? You get sucked into everything, networking, security, performance, all that stuff. So I really enjoyed that and I think that that set me up for what I’m doing today, which in 2012, 2013 I left Microsoft, I went to work at AWS and I’ve been just strictly doing cloud ever since. And it’s been kind of like, amplified my career another time. And so that’s kind of my backstory. And today I run a company called CloudSkills.io, I have a podcast of my own called CloudSkills.fm and just I’m working with customers on Azure and AWS projects, hoping to add in some Google stuff next year. That was something I wanted to do this year didn’t happen. So that’s kind of how I came up in the industry and that’s what I’m working on today.

Nicolas Blank:
Let me ask, where did the Amazon experience come from, because you were traditionally a Microsoft guy?

Mike Pfeiffer:
Yes.

Nicolas Blank:
So was it customer demand or how did that happen?

Mike Pfeiffer:
I never told this story before, so I’m excited that you asked that because I’ve never gotten into it with anybody, but … so there was a guy at AWS, his name was [inaudible 00:05:35] and he was a 20 year Microsoft veteran. He was like a principal program manager. He went over to AWS as a solutions architect. It must have been 2011 or something. And he was just a regular ISA helping people with cloud projects, but he started building solutions, white papers with infrastructure template, package it all up, give it to the customers. That got really famous within the Amazon world and he spent a couple of years building a bunch more and they became so popular that AWS is like, let’s build a team to do this. Let’s get a bunch of people, let’s make a big effort to push this up.

Mike Pfeiffer:
So he was Microsoft focus, right? So his old kind of thing was how do you run windows server on AWS? How do you run Microsoft SQL? How do you do SharePoint? How do you do Exchange? And he was the guy that was saying here’s how you do it, because he understood the AWS side, he was a real good Microsoft expert and so he closed that gap. But what happened was once he got tapped on the shoulder to go up to the senior leadership at AWS, there was an opening. And so I took his job, we built a team, and that team is still going to this day it’s AWS Quick Start and there’s hundreds, I don’t know if it’s hundreds, dozens, tons and tons of reference architectures of just common enterprise workloads. Everything’s kind of packaged up real nice together, good architectural guidance, some automation. It’s just a nice solution.

Mike Pfeiffer:
And so it’s very much like Azure Quickstart, which I loved when they came out with that after we created AWS Quick Start because it’s kind of validating that. Hey, we’re doing something that people like. But that was my in and I think for me, honestly, probably one of the biggest things going through the interview process was when the hiring manager saw that I had written a PowerShell book, which was Exchange related, actually. He’s like, “Man, you wrote a book on PowerShell?” That wasn’t just a free path to get in because I don’t know if you guys are aware of the Amazon interview style?

Nicolas Blank:
No.

Mike Pfeiffer:
It’s pretty hard. So it’s like they do multiple phone screens and then you have an interview loop, an all day interview with like six, seven different people from different groups inside the company and they beat you up pretty good and make sure you’re a culture fit. And if you’re not, they send you packing. And I know lots of people that were qualified that didn’t get in just because of that whole process. So you can be awesome and still not make it in because of lots of variables. Sometimes they’re not just technical, right? So anyways, it’s not easy to get in. So I got lucky I think having that experience. May be not lucky, but that was a good positioning play for me as a PowerShell expert. And they knew that they needed help with automation, mainly in the Microsoft space.

Mike Pfeiffer:
And so that got me in the door and then once I was in the door and I learned AWS and just started geeking out on it. So that’s how I kind of transitioned out of Microsoft stuff. I mean, I still do Microsoft stuff all the time, but that’s how I got out of the traditional Microsoft infrastructure which I was doing 15 years before that basically.

Warren du Toit:
You see that’s the word you’re using there and it’s like traditional, right? So I see this happening quite a bit is you’ve got these diehard open source students that will only use AWS. Even though Microsoft is the number one open source contributor on the planet, they ignore this fact and you’ve got this sort of cult, almost hippy following behind AWS and what it stands for, even the names they use and those kinds of things, which guys are very, very attracted to. And because of that they sort of shift their mentality to the Microsoft way.

Mike Pfeiffer:
Except with the visual studio code. Visual studio code that’s for sure.

Nicolas Blank:
Oh really?

Warren du Toit:
No, no, no, for sure. I mean code has taken over, but it’s an electron app and you can run it in Linux and life is good. So it’s more like when you have your clients and your clients are saying look, we want a strategy of some sort, whether it’s multi-cloud or not, how do you normally break that gap? How do you normally get past that? How do you show them the Azure Light or show them the AWS Light, because obviously let’s try to remain neutral.

Chris Goosen:
Or should you, right, and I think that’s the reason we are here.

Mike Pfeiffer:
I would say yes and that’s how we handle with our customers. I don’t go in there and presume that Azure is the best play for them because it might not be. Their team may be experienced with Linux only and have zero Microsoft background, but that’s just one variable. Maybe they’re trying to move a SharePoint farm into the cloud, right? On traditional BMS, there’s traditional again, but you know what I mean? Maybe they’re trying to take something from the Microsoft space that’s a better fit in Azure. Maybe they’re trying to do something very progressive with active directory and identity, right? And like obviously Azure is going to be a better play for that. So I’m always trying to gauge like how we set them up for success and not play politics.

Mike Pfeiffer:
I don’t care if they pick one over the other, if it makes sense based on the skills that they have based on the team, based on the goals going forward. I don’t try to get religious about it and I won’t work with customers that are … I generally don’t have a problem with it, but if I had a customer push back on my recommendations that not much, there I’ll go, what are you hiring me for then? Right? And I have no problem at this point being picky with who I work with. That’s why I run a small company. I’m not trying to build a Microsoft partner that’s doing all this crazy stuff. It’s more about me being able to do the projects I want to work on and actually add value. So for me that’s the measuring stick is does it make sense for the company? Can they even pull it off? Just because I’m telling them what to do doesn’t mean their team’s going to be able to do it.

Mike Pfeiffer:
Now that’s a variable that you got to think about. So that’s kind of how I try to gauge it. And what I’ve noticed is that people are kicking the tires a lot on both and then that helps them gain context and then make a decision. So I haven’t had too many yet that are religious about it, but that’s kind of how I handle it. And when there’s a bias, those are the things that I kind of push against.

Chris Goosen:
So the question I have is, I mean all the situations that make sense to have both, right? So the trend that I’ve seen with the customers I work with is there is a little bit of that religious like, oh well we’re a Microsoft shop so we’re going to go down the Azure path or AWS, maybe they’ve been doing it a little longer in terms of cloud adoption. But then there’s also the situations like mergers and acquisitions where you’ve bought a company that is now fully AWS and you’re on Azure. Do you need to go and take all that stuff and just shift it or are there scenarios where multi cloud works or even make sense for a customer? Or is it very largely dependent on what it is they’re doing?

Mike Pfeiffer:
I think it is. I think it’s 100% dependent on what they’re doing. I don’t think multi cloud just for the sake of it makes any sense.

Nicolas Blank:
Like a risk mitigation exercise or even a commercial pressure exercise.

Chris Goosen:
And that’s where I was going to go with that, right, is this-

Mike Pfeiffer:
Absolutely.

Chris Goosen:
I’ve heard customers say to me that we’re exploring both options because we don’t want to put all our eggs in one basket. But let’s be honest, this is a very big basket, right? And the basket has a lot of safeguards and mitigations is built into it. So does that argument even make sense?

Mike Pfeiffer:
Yeah. I mean, the way I look at it is this. You don’t need to go multi-cloud until you’ve gotten good with one. So if you’re not good yet with the one, which most companies aren’t at this point, I don’t know about you guys’ customers, but none of the ones I work with are phenomenal at cloud.

Nicolas Blank:
What does that mean to be good?

Mike Pfeiffer:
Well, number one, not to overspend. Every customer that I go in and work with they need help and that totally makes sense, but they’re all overspending, right? And so until you’ve got that figured out, don’t go off and do another cloud.

Chris Goosen:
It’s definitely a different mindset, right? Like even in the productivity arena where historically you built an Exchange server and you paid for the hardware that it was living on end and then you just spin up more boxes as you want, you don’t care, right? So similarly, if you have this virtual environment of VMware infrastructure where you just burning VMs up all the time, who cares? It’s a VM, right? But in the cloud, you can’t do that because every time you do that, the dollars are just going ching ching.

Mike Pfeiffer:
Yeah, you’re hearing dollar signs.

Chris Goosen:
Exactly.

Mike Pfeiffer:
I mean it’s so true. I remember the first company I ever helped set up VMware for and this is like 2007, they were all physical servers, right? And they pushed back against the ends, history repeating itself, right? We set up clusters of virtual machine or vSphere clusters and all that stuff, right. Next thing you know, like 500 BMS for a place that has 500 employees, right? So just to your point, like once you have that thing it’s expanding stuff up but the cloud is the equalizer. Like you get caught immediately in the cloud and so you just overspend. So most customers that I’ve worked with are not operationally efficient yet at one cloud. And so until you are, going to a second one doesn’t make sense because let’s face it, there’s not great management tools that kind of cover everything, right?

Mike Pfeiffer:
And you kind of need to have some kind of system. I think it’s too early for people like from 99% of businesses to be effective with multi-cloud. Maybe if you’re like storing offsite backups and want to doing everything else in another one, but like literally spending on architecture across both clouds. I don’t even know like who would do that. Netflix maybe, like most people aren’t going to be able to pull it off at this stage based on the resource that they have available to them and so multi-cloud to me is like way too early of a conversation for most customers.

Nicolas Blank:
While you mentioned the friction there is cost or cost management and operational management. Then let’s assume we’re good. Let’s assume you’ve got a traditional AWS customer, okay, and they kick the tires on IAS and they became more mature and followed a services model before infrastructure eventually. So we’ve got the whole app modernization story going. All right. Why would they consider Azure or why would a mature Azure customer consider Amazon?

Mike Pfeiffer:
I love this question because three years if you were to ask me this, I would’ve said for sure like everybody should go to AWS, that’s if they’re going to do production workloads and I don’t feel like that anymore. I think right now what’s happening is Microsoft’s doing a really great job with governance and policy and role based access controls and all of these things that we basically got with active directory. And you’re going to get that now in Azure, that’s going to change the game. Like once Azure policy is like a little bit further ahead from where it is now. I’m not saying it’s not good enough, I’m just saying once it’s even more in depth, like enterprises are just going to flock to it because they already got AD anyways. But I think long-term, you mentioned AWS was early, right? That helped them in the beginning but they haven’t been building software since 1975. Microsoft has and that matters. And so like it’s just going to catch up. Like they’re going to close the gap and in my view, based on what I’m seeing and all the services and stuff, they’re probably going to pass AWS at some point. That’s my view right now.

Nicolas Blank:
Is it fair to say that from a management and governance point of view they have, and I’m going to pick on portal as one, right? As opposed to I have a storage console, I have a nitro console, I have an iOS console. I have all these different consoles and I have different entry points and different languages for different things and Azure arguably have got one portal and one language.

Mike Pfeiffer:
Right. Yeah, that’s going to really help Microsoft. Microsoft’s awesome at building management tools. They always have been. They’re awesome at building user interfaces. Like I said, that’s going to catch up. That’s going to matter. If you look at what AWS is doing in their developer tools teams, right? If you look at their builder tools, it’s like nine different services. Like all the CICD stuff, even their command line interface stuff. It’s not even close right now to where Microsoft stuff is and so yeah, AWS was early on virtual machines and object storage and things like that, but they weren’t early on most of the other stuff, like they’ve only been building software since 1995 or whatever. So again, I think that that’s going to make a big difference.

Mike Pfeiffer:
And I think that that’s going to be what sends Microsoft over the top going forward. And like I said before, most people right now have active directory in enterprise space. AWS got a lot of traction in the early days from startups and non-enterprise type companies and they’ve been trying to win the enterprise over the last couple of years. They’ve done a decent job but management tools matter, tooling matters. Like if you just look at their developer tooling, expose your studio code it’s good and it helps people get the job done. And having multiple consoles is definitely-

Warren du Toit:
And it’s strange, right? Because when you think about the history of Amazon, okay, and you look back at Jeff Bezos and he said, I need you to make the least amount of clicks for somebody to purchase something. Okay. And the guys came back and they tried for, I think it was 12 months or whatever the hell, whatever the story goes, and they got it down to 12. 12 clicks to get somebody out the door, credit card, the works. He said no, it’s not enough. One click. One click and you can buy something. Now you know Amazon, you can. One click and you’ve got it bought. So you take the same principle and apply it to virtual machine. Or am I not seeing this correctly?

Mike Pfeiffer:
That’s an interesting point, right? That’s because Amazon’s culture is customer obsession first. That’s their leading … that’s a guiding principle. Microsoft could arguably be called that, but that’s not their core guiding principle. They’re competitor obsessed, right? Lotus notes, right? NetWare. They’re always looking at what works and then like taking it to the next level. Amazon’s good at focusing on customer experience. So in your point of retail, one click checkout, cool. But if I’m an infrastructure person, that’s what was good about EC arts or still what’s good about EC2, right? Is you can go deep and as an infrastructure person, like if I’m thinking about being customer obsessed to an infrastructure person, I want to give them all the bells and whistles that I could come up with. But I think we’re getting to a point now where it’s like it’s just a lot, right? It’s a lot of information and now we’re getting to a scale issue where it’s like, how do you manage all this stuff now starting to grow and it’s interesting time.

Chris Goosen:
And I think you made a good point or an interesting point about the traction that Amazon gained early on was with these sort of hip startup companies or companies who are not necessarily enterprise, right? Because they had developers who are potentially building stuff and they didn’t mind having a platform that they could build their own stuff on, right? Now I have a customer or I had a customer, they’re still a customer I guess, they were on Google for the longest period of time and they absolutely loved it because it wasn’t polished for them, right? They had the service, they got the email and whatever they needed, but all of their integration that they had, all of the provisioning, deep provisioning, all of that stuff was custom built by them. They built their own stuff on rails to do that because they could plug it to all the APIs.

Chris Goosen:
It was really sexy when you looked at it and you thought, man, this is great that you have a whole in house development team that can build the provisioning portal that you need. But most customers don’t have that, right? And I kind of see AWS being similar to that in that if you have the knowhow and the knowledge to be able to go, I want to build a platform that provisions machines for myself. I don’t care what your interface looks like, I’m going to redo it myself anyway. You have access to all of those APIs and all of that kind of stuff and you can go have at it. Whereas Microsoft is saying, we like to wrap it in a really pretty bow for you so that you don’t have to worry about the management piece.

Mike Pfeiffer:
It’s a good point. I mean it’s really true. I believe that Amazon stuff is deep because of that purpose, right? They want to give those hardcore customers the tools and services they need, but it could be a huge hill to climb for most enterprises, especially when they’re not super experienced.

Nicolas Blank:
Just define that for us. So we understand for example, if you’re looking at a virtual machine, I’m going to argue, and Warren’s going to correct me, that if I’m going to look at what level of control I have, once I go down to the resource template, there’s not much I can’t do, right? So what do I lose in Azure versus EC2?

Mike Pfeiffer:
I think we’re getting to the point now where there’s not much, right? For a long time there was a lot you could do in EC2 that you couldn’t do in Azure virtual machines, but that’s starting to change. And so I’ve been impressed with the Microsoft’s doing with virtual machines is like they’ve added a lot of management capabilities and stuff like that. I think one place where they’ve closed the gap recently is like whit the image marketplace and maybe the image capture process. So I got a lot of customers that still, I mean we pretty much work on deployment automation. I’m still doing-

Warren du Toit:
It wasn’t very easy.

Mike Pfeiffer:
It was definitely not easy and I mean it still could use some spit and polish. AWS has experience for that. That’s just one small place where EC2 is really slick. The ability to create copies of your virtual machine or create images, copy those images around easily to the different regions and stuff. There’s things like that under the hood that are really good but Microsoft will get there. So I would say right now there’s not a huge thing for me but I think there’s a big difference where I would say, oh, EC2 is so much better over here. Auto scaling in EC2 is pretty selective. It’s been around so long, it works really well, it’s fast. So I think if you’re thinking about auto scaling virtual machines, probably right now AWS, auto scaling in EC2 is probably a little bit more mature than virtual machines scale sets and Azure. But I don’t think it’s going to be long before Microsoft is competing in every single category in virtual machines just as good or as easy to. I mean there’s not that big a difference at this point.

Warren du Toit:
But then I guess if you think about, we can always turn it around. So take your question, flip it around. So you can go deeper into EC2, you have more customization on specific pieces, however that also leaves more opportunity from stake, right? So let’s say provisioning a virtual network of some sort, let’s say in Azure. Azure is going to say to you, we’re locking this stuff down for a reason. There’s something that you can’t get to because of X, Y, Z. Would Amazon do the same thing based on that principle? So the idea is Microsoft’s giving you guardrails so that you don’t have a problem later on. I mean ideally your security, your edge now lies with identity. I mean in some way shape or form its identity.

Warren du Toit:
Your databases are on the cloud, they are publicly accessible and so the person then becomes your edge, the person who’s consuming your product. Okay. And the old traditional firewall, port blocking, whatever the kind it is. And so from an Amazon perspective, how do you maintain that identity as opposed to in Azure when you know exactly what that identity is, that identity is uniform across the grid.

Nicolas Blank:
I’m going to argue that the killer application for Azure is Azure AD.

Mike Pfeiffer:
Yes. I mean it is the ultimate like upper hand Microsoft has right now 100%.

Warren du Toit:
And compliance wise. I mean like I don’t know how many certifications Amazon does.

Mike Pfeiffer:
Amazon’s got tons and tons of compliance certifications, they do really good there. But I think what Nicolas was saying is spot on. I mean Azure AD that’s a big difference. Azure Policy is going to be a big difference. Like, imagine 18 months from now, what you’re going to be able to do with Azure Policy that you won’t be able to do at all in AWS. That’s going to matter big time.

Warren du Toit:
But you see it already. I mean you’ve already got a really bunch of angry Dave’s at the moment because they can’t do things in Azure because policy is stopping them from doing it, which is great because they shouldn’t have access in the first place.

Mike Pfeiffer:
Correct. But here’s the thing that I’ll push back on that at least to make sure I’m clear on with that because I fight this battle all the time. You can’t not give them access, you can’t not give developers access to the cloud in the next spectrum to understand what’s going on in production. You should give them a sandbox, their own subscription where they can go in and you need to empower them with the architectural diagram and everything and templates, automation for the stuff we have in production so they can at least … if they can’t bring it up in a dev environment, they can bring it up in their own environment and understand what’s happening in production because there’s way too many companies right now that are just locking the devs out and expecting them to check code into repository and have a release pipeline bring it into production. And they have no perspective on what they’re supposed to be doing or they don’t even understand like the technologies involved.

Mike Pfeiffer:
So I 100% agree developers should not be touching production resources and we have tight controls for that in almost every cloud we want to go into. However, the expectation of just thinking somebody is just going to magically know or he’s going to write good software when they don’t know how the system or the platform works. That’s just not going to play out. It’s not going to work out.

Nicolas Blank:
The best practice around this from an Azure point of view is to have multiple subscriptions.

Mike Pfeiffer:
Yes.

Nicolas Blank:
And the dev subscription should be a free for all within reason, right? Because you don’t want them to bankrupt the company.

Mike Pfeiffer:
That’s right.

Nicolas Blank:
So there should be a financial constraint, but not necessarily an IPL constraint. And then with that you’ve got QA, which is much more restrictive. And then you’ve got prod, which is super restrictive. So ideally a developer should not be deploying or writing code into prod because your dev ops process should mature your code pipeline to deploy.

Mike Pfeiffer:
Your pipeline should be doing all of the work. Yeah, absolutely. And then you’ve got all the controls in there and you’ve got all the policies, you know what’s going on, you’ve got a history. You could go and see the audit trail. But yeah, and it’s interesting too with Microsoft shops, developers usually have an MSDN subscription, so a lot of times they do have their own. But there’s also lots of cases where I’ve gone in and the management is not allowing the team to test. You have to experiment in this world and if you don’t have an experimentation culture, you’re going to spend too much money.

Nicolas Blank:
That is on premises cultural mindset, if I can call it that. That my SAN’s expensive so don’t do anything that’s going to raise the IOPS. So no, you can’t play.

Mike Pfeiffer:
But what did you guys used to do when you were getting ready to like go to the MCM before you would go up, right? Like you guys would go in and build a huge farm or like a huge forest and you would like beat the hell out of that thing before you ever showed up in class, right? And I mean the same thing. And so like we have to be able to do that, right?

Chris Goosen:
And I still work like that though, right? To me it still is … I mock up environments all the time and I think most people on my team think I’m just ancient because I have a massive VM lab that I booked to the phone because that’s how you do it. I had someone ask me to write a script for them the other day to do something and I was like yeah, but I don’t have access to that thing, right? Like I can write it for you, but I can’t test it if I don’t have access to that thing. It’s a very difficult situation and we see this all the time with office 365 with customers wanting to bring up all these elaborate solutions in their production tenet. Yes you can be on different rings for the users to see what Microsoft are doing, but what if that integration you want to do, you should be doing that in some sort of test environment so that at least you can see what’s going on. And it doesn’t have to have 583 licenses or 585 licenses. You just need one or two of each just to see.

Mike Pfeiffer:
People are overthinking some of that stuff. They’re overthinking it and they just need to like experiment and have a learning culture and be open to change because that’s the game that we’re all in, right? Especially now when you can just move so fast. So I think that’s where a lot of people, or a lot of companies that I’m working with are struggling. It’s more a human issue or behavioral thing or a culture thing than it is a technology thing in a lot of cases.

Nicolas Blank:
Can we ask you to talk to the professional, the human, someone who is in the Amazon world and is considering the other cloud? Things are quite religious, normally. So other than the Amazon religion or the Azure religion-

Warren du Toit:
Tell him to shave his beard and to take a shower first and then make the decision.

Nicolas Blank:
He probably was listening to this podcast.

Chris Goosen:
And it’s spoken by the one guy with a beard in the room.

Nicolas Blank:
And we’ve got the applause in the background. So let’s talk to that professional. Bear in mind that you’ve got the develop experience to be able to talk to this topic. You’ve got the infrastructure experience, you’ve got the consulting and the management experience. You are an Amazon professional. How would you even consider retooling or should you consider retooling towards Azure?

Mike Pfeiffer:
Well, here’s the good news. If you’re good with AWS right now, moving into Azure is going to be a lot easier than it is for someone that’s brand new to cloud because it’s all the same stuff with different names, right? And that’s even to the point where we were kind of talking before the show, like talking about career transition and kind of moving from Exchange to the cloud. All of this stuff that we did in the Exchange world is like stuff I started doing in AWS, like building DNS with volumes and networks spanned across different regions and doing DNS. All this stuff we’ve been working on I was doing not in AWS but so what I’m getting at is like number one, it’s going to be an easier transition because you already know a lot of the patterns and practices.

Warren du Toit:
DNS’ a VM.

Mike Pfeiffer:
Yeah, object storage is basically the same concept in AWS as it is in Azure and so on, right? But the other thing that I would say to all that is like right now if all you are is AWS, that is the biggest vulnerability for your career. Because I don’t know about you guys, but I’m doing Azure and AWS and I’m going to start doing Google and I know that you guys are doing that too. Everybody’s going to start doing that. So if you’re sitting there by yourself just doing one, I think you’re going to be screwed later when you have leverage in the job market to go pick the job that you want, it’s going to help you to know more than one.

Nicolas Blank:
Well it’s like virtualizing and not knowing VMware or not knowing Xen or not knowing Hyper-V. I think that the days of a deep solid professional are numbered. Like if you do networking and it’s only Cisco, how relevant are you?

Mike Pfeiffer:
It’s so true. I have a buddy that’s a CCI a long time IT guy like us, a couple of decades and he’s like, dude, … He’s still doing the tech stuff. And he’s like, “Man, they’re turning me into a software engineer.” And so yeah, to your point, like that’s the direction this is going. So get on the bus. That’s what I would say to somebody on AWS. Don’t put all your eggs in one basket. Don’t be a one trick pony, right? Because at some point that’s going to come back to haunt you a little bit, right? So you’ve got to look and see what people are using. To me that’s why Google’s interesting. Not because I have a passion to go learn another cloud, but just because I need to be able to speak to it, because my customers are starting to ask about it otherwise I’d ignore it.

Nicolas Blank:
Contentious question. Do you think it’s relevant or do you think it’s a me too?

Mike Pfeiffer:
Do I think Google’s relevant?

Nicolas Blank:
Yeah, GCP.

Mike Pfeiffer:
I do in the big data space especially because they’re good at that and they’ve spent the last couple of decades crushing it there. So they got a lot of good tools and services in that space. And I think that for them it’s kind of like they have to be in that game, right? Because they’re data center experts, like they have to play that game to keep enabling the other parts of their business. They have to be good at building data centers. Can you be good at that? So I think they kind of have to, but I think there is an element of we want to get in that game too. And so I think it’ll be interesting for people that are building applications. I think for enterprises that are like moving, lifting and shifting and they’re already on Microsoft and it’s going to take them a little bit longer. Like obviously Azure is probably the best bet in most cases and I don’t see Google being to me like something that you’re super interested in right now. I mean I know that I need to understand that world, but I don’t envision a time where I’m going to be telling my customers, you guys got to go to Google.

Nicolas Blank:
That’s interesting. Just because we are coming towards the top of the hour and we’re going to be losing our venue

Mike Pfeiffer:
Is that already [crosstalk 00:34:43]

Chris Goosen:
And you know the thing is there’s so much that you can unpack on this conversation, on this topic.

Mike Pfeiffer:
We could keep going for a couple of hours.

Warren du Toit:
We can do a part two.

Chris Goosen:
We might have to do a part two. Maybe we need to come on your show and do a part two.

Mike Pfeiffer:
Well here’s the thing, we could do that tomorrow if you guys want. We can talk about that, but I’m going to be recording a lot here this week.

Nicolas Blank:
Let’s just ask you, can you do a plug for your show and anything else that you’d like to blog, Twitter account, LinkedIn, Wave, anything.

Mike Pfeiffer:
Thank you for that. Yeah, so definitely check out my show CloudSkills.fm. So I’ll have you guys on there and we’ll kind of do that. So I do a weekly podcasts.

Warren du Toit:
How do you keep up the cadence on that?

Mike Pfeiffer:
You got to carve it out, right? You’ve got to go on the calendar and then like, I’m kind of like that now because I’ve got so much going on in business and life and stuff, like everything’s mapped out in my calendar. I don’t do that, I have somebody do that for me. And then whatever it says on my calendar I do it. And so that’s how I’m able to stay weekly. I thought I was going to miss it last week. It would’ve been the first time we didn’t do, since we started, we didn’t do a weekly show, but I made it and published on Friday. So it’s not easy.

Nicolas Blank:
Well done.

Mike Pfeiffer:
So that’s something to check out, CloudSkills.fm. Just head over to browser, you could find it on all the other like iTunes or Apple podcasts, whatever they call it these days. Cloud skills.io is my company. So we’re 50% training, 50% consulting. And by the time you guys are listening to this, there’ll probably be something new coming out that we have. So we’ve been bringing some immersive programs over the last year. So last year we did one where it was like a six week program. It’s really inspired by Microsoft certified masters, right? And so we basically, except for going to Redmond and then getting your ass kicked for three weeks straight, right? So it’s more like jump on the zoom call once a week for four hours. So it’s cool. And then what happened is a few of those people in that program ended up co-authoring a book with Tim Warder and I. So we’ve got a book AZ 300 book coming out, Microsoft press book coming out this month. So that was really good. But anyways, we’re doing another one of those and I’m opening the doors in the end of this month, November. So it’s going to be dev ops focus. So anybody that wants to get dev ops certified in Azure.

Warren du Toit:
I’m trying to write my exam here, but the slats just broken.

Mike Pfeiffer:
Yeah, so that’s what we’re doing. So CloudSkills.io. If you go to the website and that’s not up yet, just opt into our email list at the bottom of the page, watch the CICD webinar that’s on there then you’ll get the emails when we go live. I’ve not been pretty loud about it, but that’s what we got going on. And so hoping to get a bunch of people certified and then once we do Microsoft, we’ll kind of pivot and do another one, redo some Coobernetti stuff, it’s going to be a lot of fun.

Warren du Toit:
Awesome.

Nicolas Blank:
Nice.

Warren du Toit:
Fantastic.

Chris Goosen:
Well, that sounds like something I want to check out for sure just for my own personal-

Mike Pfeiffer:
Yeah, and that’s one of the things we do in those programs. Like it’s not just me, like we’ve got all kinds of other people coming in that are experts teaching, this last run we did we had guys like Dan Wallin who’s a Docker Captain and Microsoft MVP, Steve Buchanan, who’s another Microsoft MVP, a bunch of people, Tim Warner. So we have a lot of people come through and that’s kind of what we do. So I’m just like kind of like an MCM except it’s not Microsoft people, it’s Microsoft community folks.

Chris Goosen:
That sounds awesome.

Mike Pfeiffer:
Thank you.

Chris Goosen:
Thank you for your time Mike, we really appreciate it. We’ve got a new segment that we are actually kind of kicking off here at Ignite as well. So something new to add to the show. We’re doing a community segment that we want to sort of start doing on every episode that we do. So whenever we publish every two or three or sometimes every month, depending on how we do it, we’re not quite as disciplined as you to getting it done every week. But really the idea here is, to everyone listening, is if you run a community event in your community, a meet up or if you’re putting on a Saturday event or anything like that, reach out to us via social, we’d love to get you on to sort of plug the event that you’re doing.

Chris Goosen:
We’ve also got a great sponsor who’s come on board to kind of sponsor the segment. We’re working with ENL software, is a leading provider of Exchange Hybrid and Office 365 monitoring and reporting. Their solutions help customers manage over 550 million mailboxes and they’ve got the stuff deployed in over 130 countries. So really, really cool stuff. Check them out. They’re our sort of segment sponsor and for today’s segment we actually have Nic tell us a little bit about cloud Friday.

Nicolas Blank:
Yeah. So cloud Friday is something that I started because I was frustrated with traditional user groups and I didn’t want to go anywhere on a Tuesday night.

Mike Pfeiffer:
That’s true. Yeah.

Chris Goosen:
I’ll make it on a Friday.

Mike Pfeiffer:
Stay online, lay in bed naked and just do that thing right.

Nicolas Blank:
Well there was clothing involved and we decided to do the stuff in business hours. So effectively what we’ve done, we use teams to glue two cities together. And so I’ve got myself in Cape Town, I’ve got Warren in Johannesburg and we’ve got the Cape Town and Johannesburg as your community.

Warren du Toit:
At both Microsofts.

Nicolas Blank:
At both Microsofts or other venues, but teams is the glue. And so we co-present topics across two cities and then we keep the stuff in a team’s team. So whoever attends physically or virtually gets access to all of the content, we record everything. We make it pretty and we publish in YouTube for now until stream catches up and then we’ll publish there. But for now, cloudfridays.co.za for our American audience is the place to go. We meet every second Friday where we can, we are selfish over Christmas we don’t. So we’ll kick off again in January, February.

Mike Pfeiffer:
Nice. I’ll have to check that out man, it’s awesome.

Nicolas Blank:
Yeah. Also, what we’ve decided to be very intentional about is we’ve taken all the presentation material, we publish it in Get Up, and we’re going to record the how to and make it an open source-

Warren du Toit:
Yeah in demos and all that kind of stuff.

Nicolas Blank:
How to do this particular style of community and open source and give it away.

Mike Pfeiffer:
That’s awesome. It’s really cool if you guys would do that. That’s really cool.

Nicolas Blank:
Good deal.

Warren du Toit:
And again, thank you everybody for listening to our podcast.

Mike Pfeiffer:
Thanks guys, really had a lot of fun here. Thanks a lot.

Subscribe to the CloudSkills Weekly Newletter

Get exclusive access to special trainings, updates on industry trends, and tips on how to advance your career in the tech industry.