Ep 37 – Bart Castle Part 1

In this week’s episode we talk to musician, artist, and Cloud Guru – Bart Castle! Bart weaves an engaging and brilliant tapestry of music, yurts, wanderlust and cloud stories. Buckle up and get ready to learn some cool stuff on this one!

Follow Bart!
Twitter: @cloudbart
YouTube: youtube.com/c/bartcastle

This episode has been sponsored by Meter. 

Go to meter.com/aone to book a demo now! 

Transcript

this is the art of network engineering podcast in this podcast we'll explore keynes technologies and talented people we aim to bring new information to expand your skill sets and toolbox and share the stories of fellow network engineers welcome to the art of network engineering my name is aaron uh you can find me anywhere books are sold actually that's not true at all um with me i have one of my favorite co-hosts and i'm only saying that because no one else is here to hear me say that and his name is andrew laptop hi andy hi hello hello um and we are running as a duo the dynamic duo tag team back again and we have a guest with us and you probably heard his name before if if you haven't you should probably download any social media app um and kind of get going with the whole 21st century thing um and he's got one of the coolest names too uh bart castle how are you sir i'm well hey guys how's it going aaron andy thanks for having me on yeah man yeah uh i was we were talking about this before we started i because i thought this was hilarious that you know we've been doing this podcast for quite some time now and i don't think if somebody's gonna call me out on this i know it but i don't think we've even mentioned the word cloud uh like not well we certainly didn't mention it in a in a good connotation i'll tell you that because it would have been andy cussing about it because you know that's just how he rolls sometimes um i actually manage networks and it's hard and cloud is part of it so yes i'll be honest it's confusing for me too like i've gone in there i have like an aws free tier and they started charging me out of nowhere and i was like wait a minute this isn't free anymore that's not cool and i wasn't even doing anything in there so i'm like ah so i can see that this would be confusing right exactly so so here we are a couple of networking dudes trying to extrapolate as much cloud information out of you as possible but more importantly uh we just want to know how does one like yourself become so proficient in cloud so we're just gonna kind of go through your story talk about what you did to get where you're at now and if you don't know um bart works at cbt nuggets he's the clown one of the cloud dudes at cbt nuggets if not the cloud dude the cloud dude and uh yeah and if you don't have a cpt nugget subscription go get it now because you're also living under a rock um but check it out i have one by the way the first time in my career i bought one and i i'm i'm in i'm in love with it isn't it i saw you tweeting about that i saw that that was good it's well it's amazing because like my new job not this isn't about me bart but i'm my new job i'm like there's six or seven different technology stacks that i got thrown into and cbt nugget says training for all of them so i'm just jumping around all the different nuggets i can eat a little devnet associate i need a little cloud and a little you know encore like it's just dabble it's amazing yeah i'm a big big fan of the subscription i don't know how i went this far you know in my career without it but it's really helpful right now well i'm glad to hear you're enjoying it so much one of the things i love about that organization in particular is that they really value this this personality forward fun learning experience and i really i think that that hits home in a big way for a lot of learners they really appreciate entertainment and learning at the same time yeah i mean you're a unique dude though too uh like we got to be honest here like i feel like most of us are like yeah i'm pretty unique but i'm i'm comfortable in saying that you're one unique cat like and i mean that in a good way too because like there's so much about you that i i don't understand because i'm not at that level do you know what i mean like i'm not there yet like i'm i'm like i'm not bart castle i'm like uh i'm like bart fortress right now like i'm i'm i'm gonna get i'm gonna get there at some point you're you're the moat i'm the mo yeah i'm the drawbridge well there's there's so many trainers right out there that we've all watched but then there's guys like bart who are like entertaining to watch you could be technical and know how to teach bart i've tried to teach stuff on my little youtube channel and it's boring yeah you know like you're i'm trying you know i do push-ups beforehand i'm like i'm going to be entertaining but i get ready i get 15 yeah i get 15 minutes in a routing protocol and i'm bored myself but like you're you're you're entertaining to watch which again that cbt nuggets formula it's it's really helpful to keep me engaged you know we're learning things that might not be all that exciting and you guys make it entertaining which is really helpful so man i i love to hear it yeah so so speaking of so you you you work at cbt nuggets now so start us well first of all where do you live so everybody knows if you don't mind saying i'm joining you all from asheville north carolina ash vegas down here in the mountains of uh north carolina yeah good time nice little nice the foothills of appalachia is that what yeah you know it's funny i grew up uh in boonesboro maryland on the appalachian trail and i still live like 10 minutes from the appalachian trail all the way down here in the south now i never go too far from it so so okay so you went to high school in maryland and then eventually moved so this is true let's just start there so so in high school had you had any sort of inkling that you were going to be i mean the cloud didn't exist and i i'm not trying to date you at this point but like all of us sitting here there's no way yeah there's do we need to define the cloud because the internet existed and that's kind of a cloud right well i would say specializing no they're very very right sure sure sure yeah yeah uh and you know well yeah i didn't really know about computers a whole lot back then i i was into video games uh i enjoyed playing things like everquest and getting online and downloading music and i you know so i was an internet kid i grew up with chat i was comfortable with that i remember dialing up i remember like queuing up downloads on napster before i went to school so that i could listen to him by the day last time i got home so that's that's my generation there um but i honestly i'm a musician and an artist that's really what i identify as first and so in high school that's what i was doing i was making music drawing pictures uh wasting time and kind of having a good go of it but um come graduation time my goals were to travel and that's what i did i went on the road i lived in a van for a while i went out to colorado went out to oregon um i worked at a hostel outside of yosemite for a while working in the bar there just having a great time meeting people uh and for me it's all about communication i love chatting with people and my mom i get all that from my mom she would talk to anybody anywhere anytime and so i just i really love it i like to approach people and hear their stories and and get that energy going between uh especially international folks and people from different parts of the country i just got hooked on that so i've been moving around a lot since high school and then after that i ended up meeting my wife back in maryland on one of the stints where i was home and i had actually met somebody who got me into lan parties like playing games on small networks so what game were you playing uh we were playing like battlefield 1942 yeah and um quake and we were playing unreal tournament oh my god this might be fun like that and like this this buddy this this bro of mine in colorado while we risk them and he's like dude we'll just build you a computer i can get you parts you know three or five hundred dollars there let's build something for you and so he showed me how to build a computer and i was like damn this is like it's like building a house or building a car which is what my dad does and i was like wow this there's parts you put them together and you do some cool shit with it so yeah fast forward i went back to maryland after that stint out west and i was like let's let's take a look at a computer program that i could get into and so i actually went to a network technology bachelor's program that was really really accelerated so they were all about crushing you through it big focus on certification and job placement at the end of it so i came out of there as an mcp i had my microsoft certification for running windows domains and i had my ccna and my net plus and my eight plus so wait you you went from wonderlust traveling everywhere to i'm sitting down in a in a in a school and getting five is meeting your future wife is that what kind of slowed you down and got your butt in school no actually i met her kind of towards the end of that phase um she was in a play with my sister in her college years and we just kind of hit it off and then i i convinced her to move out west to me so i went out to uh with to oregon immediately after graduating from that program um went to oregon and i got a job for lithia motors where i was a help desk technician it was my first my first tech job ever um but lithium motors had a big old fat fancy network that was national um car dealerships they have a lot of interesting use cases for tech we had a call center we had all the enterprise support infrastructure that went with that so i cut my teeth there and that's when i fell in love with networking really i mean i got into the net plus and the ccna but it was only because there's two programs the picture was either networking or software development those were the two courses tracks that they offered so i got into networking and um fell in love with voice over ip so i started i did cisco voice migrations for them and took all of their copper down and put them on a massive mpls network running cisco call manager back at 4.0 uh wayne back when calm angler first came along and this is kind of my best story about that adventure i remember being at a help desk closing tickets slurping down what were they i think it was balls that guarana drink yeah yeah i love that stuff take it back and i remember seeing um the adp techs these are guys from adp that worked for us and they came in there and there was this one dude he didn't wear a suit he didn't have the tie on that we were all wearing and he's over there mumbling about call plans and dial processing and spinning calls out this way and tail end hop off and i was like i want to be that guy i want to come into companies like this and be that guy right there so i went i started sitting next to him as often as i possibly could and was like dude show me what you're doing how does this work how do i get in your position and wouldn't you know it uh within six months i was the ip telephony administrator first ever at lithia motors and so i took on the entire enterprise network there at like i was 25 at the time and just started running the thing and it was such a great experience call center work supporting agents supporting all the crazy calling paths that you see in a car dealership support sales the sales floor the bullpen managers um ivrs that we ran so they had all the cool stuff it was a great place to learn horrible place to work um so that was it i got skilled and then i was like all right let's go find a cool place to work now that i've got some chops and i also had a whole bunch of cisco certifications by the time i left there so um fast forward after that i moved up to portland with my wife we had gotten married and we were looking at what to do next um ended up getting into a little bit of a scuffle leaving that job and landed with an ivr software development company in portland that wrote voice telephony response software and i became their network admin which meant that i was supporting their enterprise infrastructure and supporting their software developers this was my first chance to see source control start seeing the pipelines that went with us on the same point i'm running sharepoint running exchange run in our enterprise phone system that we had there as well it was a staff of like 40 people so cool little shop where i could wear a lot of hats yeah that's always good yeah okay can you explain what ivr is to everybody so if they do not know sure so that would be like interactive voice recognition or voice response there's a couple different ways you could spin it but basically these are the touch tone menus or voice menus where you would call and say press such and such for support or if you want to check your balance um and we did uh voice systems for um half of the country's state government so if you were ever in in fact right now the company still runs it uh if you're in washington state and you're going over any of the mountain passes and you want to call for road information we write that system so it checks the weather tells you what the what the passes look like it was cool it was a lot of great integration work which i think is an amazing way to learn a whole bunch of different technologies yeah so kind of like running our companies networks getting into other people's networks solving their problems getting upset about oracle learning you know what sun and solaris was and all these other crazy pieces of infrastructure that i had never touched before i just kept just gobbling it up as much as i could it's a good game curiosity it seems to be your one of your strong suits and it's funny because this gets echoed i feel like almost every time we talk to i mean andy i don't know just about everybody where the the story is like i saw a dude or i saw a lady doing something and i wanted to i wanted to be that person like i don't think a single person has not said that up until this point you know what i mean it's weird when you think about it to me because like it's almost like it's weird because it's like hey that's kind of like jealousy in a way you know what i mean it's like it's like i want what that guy has like keeping up with the joneses type thing right but totally it like i want to aspire to be that but it's weird because we all get into these careers and then we see a person that's doing something that we had no idea was a thing like i the thing that comes to mind for me was um andy correct me if i'm wrong here was like man sewer story right uh i think he said a similar thing where he saw a guy doing stuff and he was like yo what is that you know i'm tr that's my best man that's a pretty good montour it's a light bulb moment yeah he saw the guy sitting at the computer said it was like the matrix and he's like what's that and yeah same kind of thing yeah but like you don't you don't know that until you know you you're exposed to other stuff like okay like voip like to your point there's only two paths like in college especially in that particular one that you went to that training facility or whatever but but there was always so much more there was the voip guy but the voip guy i feel like especially up until that point actually up until let's just call it yes but hold on let's just call it what it is because up until the last i want to say up until maybe like five years ago those dudes they were a a group of folks that uh it was like a brotherhood like a sorority you know whatever you want to call it no one knew what they did everybody just paid them and and just kept their mouth shut you know what i mean they're like oh god the phone system's broken he you know call al you know an al poor guy he's like 70 he's trying to retire but he's the only one that knows how to work the phone we all know the the the scenario oh yeah i have images right now playing in my head say same same and it's like it's always the same thing but you were like to me you know like like i'll just give an example like when i was a cable tech i was i would see al come in and i was like dang like i thought my job was cool because i'm like giving these people internet and oh i'm also giving them dial tone but i'm at businesses right and they got to call a guy who's clearly much smarter than me to come in and type in something and i'm like hold up this dude is like a grandpa no offense by the way like i just i just want to clear that if you're a phone guy for one and also if you're 70 and you're listening to this i'm so sorry but uh but i i was like no way this dude's smarter than me you know he i you know i i'm just looking at him and i mean this is just awful to say right but it's like i mean i have i have some sort of like feeling that like hey i'm a hip kid to your point i go to lan parties i set the lan party up i know what i'm doing over here you know what i mean and i'm like i could do that i i i it was it was less like a jealousy and more like like kind of like if this guy could do it i know i can do it type of thing do you know what i mean and so i feel like we've all kind of been in that situation at one point or another but for some reason the voip the voip people and it was even before voip right like it's really telephony in general some of those patterns yeah for sure like especially key systems and especially even just like you know 66 blocks and you know toning and you know all that kind of thing um it's just a yeah right it's just a rare a rare breed of folk for sure and and i mean that in a good way by the way because those that still do it like it's it's only getting more complicated and more complicated like if you want someone to figure out like a puzzle go find someone that does like call paths and uh sets up ivr type situations like you were talking about and groups and all that kind of stuff and it's like there's nothing those folks can't figure out because to me that's harder than like routing and switching like there's just so much more that goes with it right and also folks are a lot more sensitive when it comes to getting hung up on or rather than like their pa their web page not loading they're like this phone sucks this phone sucks like whoa dude it came with a high expectation of quality all across the board yeah right right stressful yeah totally oh let me get my butt set yeah right over here the image that comes to mind is is when um jim carrey's on top of the telephone pole and the butt set's like swinging you know what i mean exactly right god i hate that reference but here i am using it um yeah but okay we're pretty cool i don't know it's all that stuff all sorts of fun things so you're your wife's an artist too right because she said she was in a play uh this that was in college she's a writer she's a freelance writer that's what she does now and she's actually a published author now she won a bunch of awards last year so i'm super proud of her yeah check out her books called the seclusion it's a little bit of a what would the u.s do if we let all of this crazy stuff go too far very dystopian um been a great show on the national indie authors award so uh we love it we got a lot of artists at home and that that's part of what's shaped our careers we're both really flexible people we don't we don't really work for other people even in cbt nuggets i still run my own consulting company on the side here and i keep a handful of clients going on that that are choice that have nice gigs and things that work well for me but right i still try to stay involved i think that's important as a trainer yeah right that's a good point right because you know you not saying that things would pass you by or pass you up but you know you do get a little rusty right like your chops i mean you're a musician right like it would be equivalent to like just giving kids guitar lessons all day but never sitting down and like jamming yourself you know what i mean yeah that's a great analogy and it's hard to do though it's hard to it's hard to come home and jam when you've been teaching guitar all day right i would be yeah yeah no that's that's exactly where i am with it right now and i i suffer a lot of wanderlust i like a lot of shiny things and i move in different directions as you might imagine and tech is always rich with places to distract you yeah that's actually a unique uh point of view there obviously you're you're keen to it because it's your life but i'm like just kind of putting two together i'm like the fact that you were just kind of like all over the place like geographically and you know you know just not necessarily like you were looking for something it seems you know it's just like you know you just enjoy being out there and seeing new things and the fact that you found yourself in tech like that that just fits perfectly because it's the same mentality it's like you won't get bored right nope there's always something else that's coming down the pipe it does not sit still yeah yeah especially in your uh discipline in the cloud which i know i still really don't know what it is um well we're getting up to that point so so the last part is i had another moment like we were just talking about sitting in a training session watching uh it's a session on avaya systems and session manager which was a sip platform that they offer yeah um learning about sip and loving it very cool i was just blown away when it came out i had been doing h323 for so long and to see set be able to normalize all of the signaling between these different controllers very powerful but anyways i saw the guy teaching and he's like talking about riding motorcycles around the country and doing fun stuff that i was into and i'm like again how do i do this i found the email the other day like where i wrote the trainer and i was like i'd love to know how you do this and he's like dude you were good in class um send me your resume if you're interested and i'll i'll hook you up so i ended up working for global knowledge uh and left that training company or left that ivr company and i went to work for global knowledge as a full-time trainer and that's where i kind of got into running my own consulting business and i grew into the cloud space as a result of them saying hey what about teaching some aws classes and some of these cloud classes we have a huge demand for it would you be into it and i got to looking at it and i was like wow this isn't i thought cloud was this weird thing that i maybe seen it on the horizon i'd heard people talking about it i had used aws quite a bit at our previous company to kind of play around with it but even though i wasn't really thinking of it as like what we think of as cloud right now office 365 was just barely making waves at the time when that was happening so anyways fast forward and i kept following my way through global knowledge getting certified in different vendors started working in the aws space and supporting some clients along the way there and then fast forward now that's been about eight years ago now since i became a full-time trainer um and i think the biggest thing to get back to like what is the cloud it is all of the things man if you're doing networking and you're working with network attached resources and systems uh that is cloud computing and i think that's one of the biggest things for me as a trainer is that a lot of people know more about cloud than they think they do you might not know all the vendors yet but you're still very familiar with the principles of automation or virtualization or wide area and local area networking these are these are the these are the underlying technologies the rest of it it's just a bunch of software and applications that live on top of it so that that for me i realized that i was actually in a really ideal place i'd run virtual infrastructure for many years on vmware and on this was before esx i was even around yet and running some of the was it i think kvm that we were running for a while there uh zen uh not citrix but the original zen kernel parts of it so doing that work and starting to transition into the cloud computing piece i realized that i really wasn't transitioning as much as i thought i was i was just working with different vendors at that point so that's one of the beautiful things that i love about cloud computing and where i am now is that it is not a piece of the technology pie it is the entire thing i mean you can touch anything you want you get into analytics you want to get into security you want to get into applications or serverless software development or if you just want to help organizations run their traditional monolithic applications then a career in cloud computing gives you a chance to do all of those things and transfer around to different companies and different industries and even different roles in those same companies um by taking the cloud avenue so it's it's kept me and continued to indulge my curiosity so i mean like saying that like you're in so okay so if i like work in an organization or whatever and like let's say i'm the the cloud i mean is there a cloud guy like at an enterprise you know what i mean like like oh i'm the i'm the cloud guy right or is it just like like you just said is it because it's like the part of the networking guy part of the software development guy like uh the short answer is yes you will see that um what i originally started with was trying to teach people how to use the tools but the longer i spend in the cloud computing career uh the longer that i do this the more time i spend talking to the business side of the house and saying you know what if that's your approach right now you're missing a large part of the conversation um moving to cloud using whatever tools you're using already you're probably just going to perpetuate a lot of the problems you have already so if you're not looking at how this can help you innovate and uniquely solve problems in a new way then it's just another hosted data center that you're talking about and you're not really talking about the growth and innovation or the changes that you might really need to make as a culture in your organization so if you're asking is it somebody that we need to create we need to hire a cloud guy you might have already missed the point of what we should really be talking about so i spent a lot of time backing people off of that ledge and saying all right let's go back to business objectives what is your key demographic what is your key uh output what is how does it drive your revenue flow and going back to the business objectives and saying we want to make sure that you're using the right tools in the right way and paying for them in the right way to get the best value for the organization so that's really where i am now is having a lot more of those cloud vision and strategy conversations with organizations that and doing a lot of security auditing because people throw stuff into the cloud and then they realize that they don't know what they're doing they get audited and things just start falling like dominoes so there's there's a lot of easy pickup work for me to get in on just explaining cloud services to their internal auditors so that they know what questions to ask their teams um and those things just kind of happened organically as a part of working with these different vendors and then working with different clients that are using their services so well so because andy's the cloud guy he doesn't know it he doesn't know but he's the cloud guy he does know it he just doesn't know what the cloud is you said a couple of things in there that made me think of andy like right away it's like like right at the beginning it's like oh yeah somebody just kind of gets thrusted into it and you know hey we're the cloud but i'll i'll use him as an example as well when because you were talking about security and audits and stuff this guy right here works in probably one of the most heavily audited secure situations because you know he does he works for a financial institution right so okay moves lots of money around right so it's big it's big stuff and the government's going to come after them and knock the hell out of them for whatever reason just because they need to pay somebody with our taxes um but that being said like was there a period there where people were afraid to go to the cloud because of that because it's scary in a way because like i think the the thought process is this it's like okay if it's the idea is that it's readily available to me wherever i am at any time and it's and it's resilient meaning that there's probably duplicates of stuff everywhere doesn't that just open the door for you know more opportunities to come and grab my stuff right sure well then you pair that up with people throwing terms like public and private cloud around like oh you can either put it in a safe place or you can leave it sitting on the street out here and it's like okay wait a minute hold on a sec you guys use public banks all the time and you don't see a public institution as a bank in that light yet that's exactly the characterization that we have of public cloud vendors is that there's this laissez-faire approach to just allowing everyone to peruse each other's stuff and it's like hold on a second yes it is multi-tenant there are many people cohabitating the same infrastructure but there are a lot of isolation and segmentation mechanisms that have been employed um for a number of reasons not just for the customer's interests but for the interest of the provider as well and probably the biggest thing that i try to encourage people to recognize here is that when you're working with and i would say with a disclaimer that most of the time when we say public cloud people are thinking azure they're thinking aws they're thinking google cloud the big infrastructure the big infrastructure class ones okay and in a lot of times if you're moving to infrastructure as a service it is not about chasing value from a monetary perspective okay it's often about perpetuating what you have already i almost always go with the rule of follow software as a service if you can't do it with software as a service you need to double challenge yourself to prove why you need to go to an infrastructure as a service provider it's a really big kind of pet peeve for me so explain that because because okay so keep in mind yeah yeah we haven't said the word cloud once in six months so so when you say think about it from a software as a service rather than an infrastructure service first let's let's just talk about what the difference is there because i think i mean i i think i know and i'm i'm going to speak for andy on he thinks he probably knows too right so collectively we think we know um but what what would what would you say okay so i i i first of all i try to encourage anybody who's getting into this field to immediately go and read the national institute of standards and technologies publications there's two major ones around it the first one is about service models and deployment models where they lay out definitions of what it means to be at the infrastructure platform or software as a service level so typically you're thinking about management layers you're thinking physical resources data centers locks doors server racks power heating you're moving up into hypervisors and operating systems you're getting into uh applications that run inside of them you're thinking about authentication and data and users that live on top of it as you move between those layers infrastructure is the one where you're basically getting virtual machines and virtual networks to go and configure to your heart's content just like if you were running it on vmware if you're going to software as a service you're getting a turnkey product that is generally going to be available over the web it could be software that you're downloading installing sometimes it's just a licensing option the point is that the administrative layers that you're dealing with are drastically reduced so you're down to the point where you're just consuming the application levels and you're not thinking about a huge section of all those other administrative layers underneath of it and that's exactly why i say start at the top if it's not sas then you better be careful about picking the right platforms and infrastructure that you choose to do battle with because the returns there are marginal and difficult to come by and they require a lot of optimization to get it right and a lot of our legacy systems they're not very cloud friendly so unfortunately i do spend a lot of time trying to talk organizations out of moving things to the cloud it's usually one of them the more important things i do as a solutions architect to say that's a good fit that's a good fit these 12 other things you were talking about keep them in your own data centers for now for these reasons wait till the timing is right if it's not high value and low risk call me when it is um because that's i mean that's kind of where i am with it i can be kind of hard because we're on this this fun environment right now but that's those are the lines that i try to draw and get new solutions architects thinking about you need to be able to justify that stuff to them which makes it a tricky field to be in if you don't understand some of the limits to your applications and infrastructure moving them to the cloud is only going to accelerate many of those problems and you could get yourself a really big fat bill at the end of the month and it's not going to be better so i really have to be careful with like where you go when you pick it up this is this brings up an interesting point then because you know this is the the bright shiny object it's still it has been for gosh like 15 years now i feel like people are like oh we're going to move to the cloud or whoa we're not ready so so that's how that's how they sit around the big round table they just shake their shoulders back and forth we're moving to the cloud this is exactly how meetings go give me a project manager stat everybody get on board we're going to the cloud so so look okay so that's probably part of the issue right because like all these people bart's talking to these are all like executives decision makers and they all go to the same meetings and read the same magazines or whatever and yeah the cloud's shiny we got to be in the cloud our competitors in the cloud we got to go there it's kind of interesting i never thought that like a cloud guy like yourself would be talking people out of it yeah so give us an example then so so if you i mean you don't have to name names obviously or or give us a specifics as to who did what but but okay can you give us like an idea of like when you would tell someone to pump the brakes and not move to the cloud i mean is it is it generally like a you guys are using this much of your data center and it's like why move that because you're not you're going to be using that much of a cloud compute space anyway i don't know walk us through kind of like a scenario if you would um so i usually go back to one of the hardest things to try to quantify in the tech space and that is total cost of ownership if you've ever been a part of any sort of like tco analysis process it is the absolute pinnacle of herding cats and gray area and nebulous divisional like uh like weird definitions of which team owns what and how that depends on another and how do you attribute that back so if i first of all talk to an organization that understands and can really navigate their total cost of ownership that is a good sign right out of the bat if they really understand how their individual business units leverage a piece of technology and they also have built the right complementary um internal billback chargeback models or financial allocation models to support identifying where money is actually going from a technical perspective those are all really good signs if you're walking into an organization who's still kind of like ah we're just putting some virtualization together it's been a good time we still own all these data centers and they're still kind of big hammering future proofing their purchases like buying the largest systems implementing the most i don't know fail-proof redundant design that they can those are really big trouble signs for me the biggest concern for me is that the business has not really enabled the technical teams and business teams to identify where costs actually live and so it's going to be really hard for me to demonstrate why we should move a particular system or service to the cloud if i can't show you the technical value of it or the business value of it so those are really the first trouble signs for me if an organization can't bring me that sort of info then we need to go back to the discovery analysis and evaluation process like tenfold a good example this would simply have to do with a lease okay you buy a server it came with a lease okay it doesn't mean that moving the systems that run on that server is a bad idea it just means that maybe the timing is not right and so if you still have a return on that investment that's happening or you haven't realized the return yet then we don't just evacuate because you're still working that timeline and we also recognize too that if there isn't enough value then we don't want to introduce additional risk so these are all things that i kind of get them talking about we contrast that against what if we could go to using a software as a service solution where we only had to move the data part of it and we could just second just kind of cut off that portion of the infrastructure and stop using it if it's a real clear cut yes we can turn it on and off that's a pretty good fit that represents high value low risk there's really clearly defined transition window opportunities that they already kind of have an understanding of those are all really encouraging signs for me and often it's not the most important applications in their environments that should be moving this is why things like salesforce crm are such a popular platform because it's it's a commodity style service that lots of organizations use and they can get a big value out of leveraging the expertise that those vendors have without having to go and skill their teams up without having to go and buy infrastructure and install it and then maintain it yada yada back to cost of ownership so for me it all goes back to tco if you can't bring me compelling tco results to begin with you should probably pump the brakes there and go into a mode where start new deployments look at the high value low risk opportunity so i would say doing pilots doing anything that is a fundamental capacity planning problem for you things are going to be extremely short-lived that you're not sure that you want to use yet run those things in cloud providers find out what is good and bad about them and turn those things off and stop using them or go into a mode where you're finally ready to do a real pilot that shows you and you can actually contrast against platforms that might run that for you running microsoft exchange is such a textbook example of this for me right now lots of people run exchange servers they've been doing it for years right but office 365 is an almost clear-cut open shut win especially when they've already addressed all of the active directory dependency problems that go into getting that authentication layer to work right which is one of the biggest hurdles early on so things like that when it's a clear path that's when i would encourage them to go to these software as a service solutions on the infrastructure side if you're not in a position where you're willing to turn the entire data center off and just commit to a giant vendor then you're probably in one of the gray areas and that's where most of the companies that i talk to on a regular basis are so those are just some simple examples and some of the red indicators that i look for when i talk to teams um if they know what they know a lot like like you walk into places and you're like nah nah because they're like hey first of all you're cloud bart you're here for you're here for something and i and i'm gonna guess that it's cloud so so so like because like just hearing you talk kind of blows my mind i'll be honest with you because i've talked to a lot of cios and a lot of ceos throughout my career and i'm gonna just throw this out there i'm willing to bet that you've walked in and given that spiel before and just completely gotten smoked to come out of their ears because and these guys and gals are you know cios and ctos and ceos and cmos and whatever whatever you want and sea levels sea levels yeah the c-suite it's not so sweet get out of here with that they're in the c basement the c dungeon um so like colorful if you're telling if you tell somebody because this is where i'm going with this so you walk in they're like all right we got cloud bark coming in you know we're going to talk to him obviously about cloud because you know it's cloud bart um and you walk in and you're like uh you guys don't need the clock like what what is happening in those conversations when you're just like nah like are they getting offended because you're like you ain't ready for this like like how do they feel about that oh yeah well you know the truth is that i i very carefully select clients now um i look for trouble signs like this there are some particular industries that are just going to be trouble no no okay like what i mean i steer pretty clear the finance uh world it's not that i wouldn't be willing to work with them i would rather help them audit and do things that are more like complementary to what they're already doing as opposed to like big picture discussions um i just always run into problems with things like that i tend to work a lot more closely with software development companies these days that's usually the ones that i try to seek out more often especially if they're smaller and they're more agile i just think in a lot of ways the wins that they have available to them and the lack of uh encumbrant that they generally represent those are good signs for me uh also the malleability of the organization so i generally look at factors like that before i try to engage with a client because i don't want to walk into rooms like that and have that reaction i've seen that happen i've heard lots of other horror stories from a lot of other peers that see that happen as well it's a great way to earn business for yourself because quickly you find yourself either kicked right out immediately or you find yourself working with that client for many many years to come because you're so trustworthy at that point right you're telling them the truth which is coming straight yeah i mean whether they want to hear it or not so i i have to ask and dig in but dig in carefully just because the industry i'm in so why why specifically do you avoid finance what what what makes that vertical such a nightmare to to work with um it's not even so much that it's just finance it's that i find that in order to uh assuage some of the concerns of those organizations it helps to contain it helps to maintain some additional certifications and visibility um around some of the reporting that has to be done some of the standardized auditing that has to be done and i just frankly don't maintain all of those pieces well enough it's the same reason why i don't work with a lot of defense organizations i don't have security clearance to work at some of those levels so i usually just kind of avoid some of the higher ability logistics of it it's not that i don't think i have something to offer them it's just that i don't really need a six-month ramp up for you know a four-week consulting gig and i've done that i will probably do some of that in the future still but that being said i've had some gigs with companies like visa and some of their card processing and payment processing divisions as well and they're great companies but i tend to at this point i'm in with them now so they can kind of slice me off a chunk with one of their different partners that they work with and there's a whole lot less startup time for me so working with organizations like that has helped but it's already because i have a relationship with them that smooths some of that entry i'm just kind of sensitive to that i'm also juggling things here so i'm trying to be realistic about what i can do i don't want to compromise quality of work with um too many too many efforts i'm a father and a parent too so i don't want it right right i'm trying to think about those things as well quality life oh andy knows all about that he you just struck a chord with him he's like how do you do it how do you do it bart teach us he's about to move his whole family he's moving his whole family to the cloud if you convince them but that being said another client that i have was another really big payment uh competitor and one of my favorites to work with they're a really high profile aws customer and those guys are amazing the way that they rolled it out uh for them they started by really upskilling the security and the network teams those are the two teams that went cloud first they went and they got certified they established standard operating procedures over how deployments would work and what that did was it made it possible for the other divisions to onboard in their model of using the cloud rather than jumping into some vendors model that they had designed and i think that's fundamentally the big problem here is with software as a service you show up and you get the thing they hand you out of the window but with infrastructure as a service you are crafting the experience from the ground up the deployment the policies the processes and if you don't have all of those pieces in line it's like walking into a giant sam's club and saying oh i guess i'll just buy a little bit of this here and there you're not gonna end up with all of these boxes of paper towels and giant chunks of orange juice and you sampled everything that you came across and before you know it you you forgot what you walked in store for and right it happens all the time but policy first and top-down envisioning of cloud purpose uh is always paired with success um when it comes to like these big migrations that we hear people talking about what do you think is the biggest you know so somebody sells the idea of cloud to all these folks right is there like a top three like in my in my mind like you go to cloud you're going to save money but i've been told by people with experience not so much right you got to watch out for that but isn't that the thing people say right like oh go to the cloud you're not managing your infrastructure it'll be less expensive but once you get in there and you're using it and utilizing it it's kind of a break even almost right um i would say that if your teams recognize where to optimize and they've done a good cloud fit analysis it's certainly reasonable um one of the biggest concepts that i was just joking about spot instances you've heard this term spot we've heard about that before no no okay basically they're like impermanent virtual machines aws is like okay so we've got thousands of customers they use stuff kind of randomly we never really know how many they're going to be needing and it's a capacity problem for us as a provider right so we're going to incentivize aaron and andy to use our surplus resource and we're going to make it available to you at a discount that's the incentive but we might need those resources for top paying customers so we're going to reserve the right to pull those resources back when we need them ah all right now now this is usually the part where everybody's like you know they're gonna they're gonna shut my servers off and it's like okay but think about some of the strange little auxiliary tasks and workloads that happen in an enterprise data center and how infrequent or how um not consistently they are executed and they don't run all the time like transactional stuff and finance is an example yeah yeah oh quarterly stuff things that come only during report times or only at business year end a lot of great examples college admissions is another one that i talk a lot about as well um anything that has a weird seasonality or the bigger term is it does it represent a capacity planning problem for you and your team if you can't justify buying and owning and running an asset for at least 70 percent of the year odds are you could probably turn it off and there would be some win that comes from using a cloud service so this is the point where i start talking to teams in the operations folks and saying well can we turn them off is it possible that we could downscale the sides what if we run a fleet of them that are smaller and you can lose some of them along the way how do we remove some of the dependence on this this beautiful thing that we call him bob he's a very nice server we've had him around for a very long time like no no no we're putting that server down and we're going to make that a task that we can run on any server that's what i want to see so that's they call it pets versus cattle so if you want to bring all your pets at you to the cloud watch out because that is going to be trouble it'll be humorous yeah i know some great server names that are out there i know we had a lot of fun name on our systems back in the day but in the end what it does is it speaks to the cloud savvy of the organization this is another reason why i look at developer shops more often because they generally live in a more impermanent state because of the rate of software development you have to be in a state of flux to keep up with a lot of the changes that are happening in the software development world especially open source so teams like that are more uh willing to say okay well what if we pull the data store out of there put it somewhere else and now we've just got these applications that are running how do we get them more portable how do we get them to live outside of the vm what parts can be lost how could we restart parts of it and as long as you're looking at the right types of applications andy's right you can save some money in the cloud but the problem is a lot of organizations don't think of it that way they're like we used to do outsourcing we know what we're doing just let us in the data center and we'll start building stuff right now and before you know it they've got a whole bunch of long-running workloads that they're paying for by the minute or by the second as opposed to buying them up front so yeah that's a horrible roi and you were definitely going to end up spending a lot more money on those workloads than you did running them in your own data centers the only saving grace is that your total cost of ownership comes as a bill from the vendor instead of having to wonder where all the costs actually live okay that's not bad but it's not it's not the value that you're looking for no yeah i know andy's right i hear the same thing it's like oh no no why would you pay for power and then you know buying a server and then you know renting the space and then the eight you know they it's funny when you talk about uh total cost of ownership it's like somebody's doing like a a revenue profit a p l or something they're like yeah they're like all right well we made 22 widgets today and it's 15 cents per kilowatt hour for power the air conditioning costs act like you're really gonna do that like each widget costs us 14 and 63 cents like like you were saying earlier it's like no one has that well not i say no one most most enterprises don't have that it's great yeah that granularity of like especially like when you talk about the the cost center right cost center being like the different uh department within the enterprise that has their own budget and all that other kind of stuff right so it's like no one is keeping for the most part that that good at track so so it almost begs the question like when someone's like okay well we're gonna move to the cloud so we could save some money you know if they would just have somebody like my wife and i watched shark tank all the time right and they you guys ever watch that such a good show yeah yeah it's i love it um but mark cuban's always like well what tell me you know what's your cost goods what's it landed you know all this other stuff and half the time people were like uh like i don't know well half these places don't have a mark cuban in there going hey what is what did you pay for that what is this guy doing you know what i mean like they're they're just kind of like oh cloud kind of like andy and i just are dude i signed up for a free aws account because i thought it was going to be the bee's knees too i mean turns out it's not free either by the way um be careful with the freezer yeah you really do uh i'll be honest the one thing i did learn and i learned really well was the first thing that i should do is go in there and set up the billing alerts and the threshold that's the world number one that's the first thing i that's the first thing i did when i started studying cloud yep yeah yeah and threshold that's good news friends that's good news yeah we did we certainly didn't want to come over here and disappoint so so you may or may not save money right so but i feel like that's the big sales pitch that that people give like from my seat bart it seems like the biggest advantage to cloud is the speed at which you can deploy something because i i can tell you working for a software company for us to order circuits get them in stand up gear get it all conf like for us to bring up something live yeah it could take six months depending on all the processes and how long taco takes and but if you have if you have a cloud presence and you want to spin up something new i mean how fa you know pretty quickly right like you can get something going on in days maybe like a new service just boop um i mean it literally could be minutes to seconds to bring things online uh when you get into like integration and back end pieces that need to talk to each other that's where stuff gets really complex which is why i was kind of saying that when organizations are looking at how to address cloud options we look at the capacity planning problem so if you are a software development shop and you have a lot of server churn that's happening those are probably pretty good workloads that don't run regularly things that can be turned on and off you're working with a highly skilled group of individuals who are comfortable running a new technology those are all check check check check good marks for me but if you're talking about trying to take this massive system that you depend on day to day week to week and don't understand all the costs and element components of it then yeah that's going to be really really hard for us to move it to the cloud and get it to run in a state which will return the value that you're looking for and on the timelines that you're talking about there um so what i love to do is work with serverless teams who are actually able to craft what they call cloud native applications this would be where you're like running an aws a series of lambdas let's be hold on hold on hold on keep in mind we're we're because i still don't understand serverless how can something yeah that's what i was gonna say yeah that's nonsense right sure sure so when we the easiest thing to like replace it with is to recognize you're talking about managed services that that's what we mean by serverless you're not not using servers clear there's compute there's network there's ram there's storage that's happening back there but the cloud vendor has enabled you through a platform to not have to focus on those layers and it can be a couple of different degrees you might be thinking do we mean containers yeah containers are part of that but there's still a lot of work to containers as opposed to looking like azure blob storage or looking at uh amazon s3 those are services where the only thing you talk to is an api on the other side of it and you can do it through the management console you can do it through a massive ecosystem of third-party products or you can write your own software to talk to those systems and you never touch a storage component you never touch a network component in the background you never deal with resource outages you never deal with the performance or scaling parts of it those are hallmarks of serverless they can scale to an extremely high degree they are managed to a high degree and it allows you to largely focus on the differentiating logic of what the application or component does and not the implementation and management of it those are those are hallmarks of what serverless is uh you don't really want to pin it down to one definition you want to think about some of the characteristics of a serverless application that that helps uh for me so api driven very key part of that super abstracted for you typically you're thinking about a series of calls that you're making to get the system to do something as opposed to running an operating system and patching it and supporting it and putting it to bed at night and buying it birthday presents you know all of those kind of funny systems operations pieces that add to the complexity of cost of ownership when s3 just charges me for the amount of space that i consume and the number of calls i make to it that is a very different consumption model than what almost every traditional ops team is used to dealing with and therein lies the problem with trying to get the right value out of cloud we're not talking the same language and we're not looking at it in the right lenses from the very beginning so it is really a very big big bridge to cross for a lot of teams one thing that i try to get teams thinking about is cloud champions one of the things that our team does at cbt is having centers of excellence around individual team members so if i have 12 software developers i don't want them all to be experts in the same thing i want to have an expert here an expert here different technology a b c d and i want them to be able to collaborate together so that my entire team works better than the individual components themselves and so having cloud champions and having people specialize in specific areas of functionality these are all good patterns that fit the cloud software development model and the cloud utilization model that it gives us a chance to consume a really wild ecosystem aws has like hundreds of products and services that they use with thousands of features you could never expect anyone to know and i i live in that space and i don't know what all right that that actually i live in my documentation abilities i mean that that is what i do i know how to find the answers not the exact answer and it's it's sad but you got to know that that's you cannot realistically take it all on that's not i started going down the aws path you know i'm like all right i'm going to get my you know architect associate and i was i was going through adrian cantrell's class from linux academy and i got about a month in and realized how dense and how deep this was and was like i don't know if i have the time you know i got three other things i got to learn so i actually bailed at that point just because yeah i'm amazed at how much is going on in in the cloud it's i don't know how you guys know it all it's amazing hey folks this is andy when you have someone as prolific as bart castle on your show it can easily turn into two episodes so that's what we're doing here this is the end of episode one and we will see you next week for part two of park castle see you then hey everyone this is aj if you like what you heard today then make sure you subscribe to our podcast and your favorite podcatcher smash that bell icon to get notified of all of our future episodes also follow us on twitter and instagram we are at art of net edge that's art of n-e-t-e-n-g you can also find us on the web at art of network engineering dot com where we post all of our show notes you can read blog articles from the co-hosts and guests and also a lot more news and info from the networking world thanks for listening you