Cloud Crunch
Cloud Crunch

Episode · 2 years ago

S1E02: History of Cloud and where we are today


When did we first hear about AWS, the cloud starters? Where have we come since then? We dive into AWS price change trends, new service launches, how Google Cloud Platform and Microsoft Azure came into the mix, and how it all affects you.

Involve, solve evolved. Welcome to cloud crunch, the podcast for any large enterprise planning on moving to or is in the midst of moving to, the cloud, hosted by the cloud computing experts from Second Watch, Jeff Aiden, cofounder and Executive Vice President of marketing and business development. Ian Will be chief architect cloud solutions, and Skip Berry, executive director of cloud enablement. And now here are your hosts of cloud crunch. Hey everybody. Jeff Aiden, cofounder, Second Watch, here and I'm joined by the cohost Ian Willoughby and skip Berry. Today show we are going to give a history lesson of cloud computing over the ten plus years, predominantly around infrastructure as a service, and a great way to start off is let's go to Ian and skip our experts and ask when they first became familiar with cloud computing, either Amazon or Azure, over their course of history. In all great yeah, they mean this is it was exciting times. It was back when I was managing data centers and building those out. A boy, you know, three am car trips to the worst parts of town or just not really the most exciting things in life. Friend of mine introduced me to Amazon websters as I got to check this out play with that US three a little bit. But when you see, too came out, there was no console. So in order to ploy things you had to do it all through the command lack. What year was that in about? And I was going to sit must been around two thousand and ten, maybe a little bit earlier, not really quite sure. At this point kind of all blurs together. Not Sleeping very much, going to data centers to late at night. So there was this plugin called the elastic box which only worked on Firefox, which was the first kind of Gooi based system, in order to launch it instances. But that power of launching a Linux machine and a few minutes and it actually workings like we got to do this. It was all over that. So before joined Second Watch, I ran a company and the telecom space. Ultimately migrate at all our infrastructure of it day, Amazon web services, and it was just so much better. The scale, the speed of which we could do it. It was just amazing. And now I've continued this journey, you know, and it's funny you said that, because I remember the first time I saw one of our senior engineers actually launch resources on a plane over Wi fi and we were chuckling because he didn't actually have to be physically there to unpack the box, rack the rack, plug it in and actually deploy services. All Right, skip. What about you? Yeah, similar event. Working for a smaller size software development company that was in the the compliance and regulatory space, working globally, and we had a challenge of where we would build solutions that it needed to be deployed before we got on site. So, if you think about where we were two thousand and six, two thousand and seven, ect came along and what it helped us do is do prebuilds before we went out. We were doing it before and Vm were infrastructure we worked. It was very practical, but very non cost effective, if you will, and so with that we were able to use ECTO to save a huge amount of cost and it just gave us this big, tremendous amount of flexibility to really meet more of a will say, a tailor design for the clients globally that we were working for and didn't have to drag along hardware. I was shifting almost like you would do in a trade show, shipping just as much equipment to a customer that was just staging to go and build stuff that we were going to leave behind and then shift the gear back. So just with the advent of the cloud and having that ability with Ectwo that came along, and then has three, when s three came along as well, that gave us the ability that basically we were using it really at the Code Bank, which I don't know if that was ever the intention of, but it helped out immensely what we were trying to achieve with perforce was the tool of the day back then for our developing efforts. But that was my early introduction and I'll tell you one thing that I said to some of my colleagues. I still stay in...

...touch with them now. You know that. I'm like, just just watch how big this gets. This is going to be revolutionary and low behold, here we are so great. Yeah, in my experiences is probably not as technical as you guys is. But you know, back in two thousand and eight when the economy kind of tanked, I was running a Pino, my cofounder was running technology, and when I was asked to grow the company by fifteen percent while cutting millions of dollars out of our budget, and we stumbled across Amazon web services and actually started doing some marketing campaigns up there. That saved US literally hundreds of thousands of dollars that we were able to push the savings and able to do more with us. And a lot of people don't know that Amazon web services was started out of Amazon. Back when Amazon had a lot of holiday traffic. They had to have a lot of servers to serve up that content for their customers. Well outside of the retail season, those servers sat segments, so to the storage and sat blank and Amazon figured out, Andy Jesse actually wrote the narrative, that they should rent out this computing and storage services when they're not using them, and that's essentially how aws started. For those interested, there's a great book out there called the everything store by Bradstone. It's great read about Amazon understanding their culture and as we launched Second Watch in two thousand and ten, that book just came out about that time period. I also think some of the other players, guys check me on this, but like azure and Google kind of really didn't start to take the clouds seriously until about two thousand and fifteen, twenty sixteen ranges. Is that sound right from a timeframe perspective? Yeah, yeah, absolutely. Yeah, and just just you know, not to be hull I believe what they were building to is what they do in a Protott was a target or some major retailer or where they wanted to basically prove this out, that they could put a market place. It was Marketcom was, I think the original name name. If I'm not mistaken, target was, target wasn't. Original customer was original customer and I think it was around two thousand and twelve that they actually tried to build their own data center. Two Thousand and thirteen and actually target left aws and on the day they left a bit aws they had security breach, if I recall. We'd have to go back and verify that, but I believe that was the case. One of those rare cases where somebody has left a public clown. Well, guys, let's talk about some of the examples of how cloud has evolved since two thousand and eight, two thousand and ten, since you've worked there it. Let's maybe start with some pricing examples. I think there's been sixty seven or more time since two thousand and six that the prices have changed. But let's talk about some of those, those examples. I think it's obviously sixty seven is a huge number. You know, we're used to things increasing in price. Obviously out there, but a little misleading because I think the number would actually be higher if you take into account the new generations of technologies that exist. So if you're looking at, instance, types, you know, they go from a generation of like a c three, c four, c five. You get so much more price performant in those new generations. So and they don't call out a price reduction. So I kind of woulds if you have kind of upgrade that processor types and those types of things, I think that you actually experience better. Thank for the book. Well, hold on the in. What you're saying there is really important. So not only have we seen price cuts versus price increases right, I mean you hit it right on ahead. Cloud computings not getting more expensive, but when they come out with a new model, like if you came out with a new tesla or a new you know Outi, you're actually saying it's actually more affordable than the previous generation. That's correct.

Yeah, that's a novel. Novel actually attribute of the cloud itself, right, as opposed to your old land based services, if you will, when Ip stand up provide infrastructure, platform, etc. You can't do that that way right. So the other angle on this too is just to understand how cloud evolved really from the pricing model just in general, where it was, you know, back from a mainframe perspective. Evolution of time based volume, based flap rate, you know, I made some notes here talk about it, just like responsive pricing, you know, the spot pricing that we see today. Those unique opportunities and understanding pricing schemes for the application of what you're going to use the cloud is ever so important and when we do stuff like Cira and what have you, cloud readiness assessments for applications, as we're moving people to the cloud, not to underestimate what that new world is going to look like so we can help clients really understand their spend as well. So the whole TCO picture comes into focus when we understand what kind of cost scheme is going to be put out there. So yeah, for those for those late adoptors out there, I think the headline here is you know, and guys check me on this, we have not seen a case where cloud costs have actually increased cloud usage as but you know, since it actually has been created cloud costs of decrease and now with competition, there's even more aggressive discounting and pricing. Correct. Yeah, the Big Three by it. Yeah, yeah, yeah, you got regalias that are competing heavily against each other and at the end of the day, the customers winning. And you know what's interesting about that, guys too, is we have seen with some of the late doctors, them still want to do the validation TCO type or Oroi type tests, and I think those have moved along a lot faster with the late adoptors because there's so many validation points than they did, you know, five, six, seven, eight years ago. The other thing that I've noticed with the large enterprises they're less sensitive to hey, is storage cheaper or is compute cheaper? They're looking for more aggregated discounts, which we've seen the providers become more aggressive. So here's my total breath and depth of usage on your platform and I'm looking for an overall just blanket discount. We've seen that change to over the last couple of years. What about global services? If you guys, you know, walk back in time, you know, we see this today with services. It's not deployed globally. But what were some of the maybe issues or challenges you saw? You know, or the growth that we've seen over the last couple of years with global services around cloud services. I think the greatest services that were starting to see are some of the on the global side. If you look at some of the database technology, where it was really localized to one region, maybe multiple zones as well, but this multimaster scenario, which is something that when I was building out my own infrastructures, is something I really really wanted to have. An acid, compliant database that spanned across geographies was just kind of like the holy grail for high availability and disaster recovery. Those things continue and if you look at not only is it great that is these global services, but it's still good that they're isolated as well so that if one gets destroyed for whatever reason, it's not going to influence them across the board. So I think it's very exciting. Obviously, invus is one that's very level, that really kind of spans across everything, and then making sure also you don't go out of control on those things as well, so kind of shutting down and making sure you don't go to global keep those cost containments in place. Yeah, I think the other side driving that to is even compliance from our from a perspective. But while so you have regulatory bodies, you know, your GDP are all that kind of stuff that says, you know, data must live x and these boundaries and what have you. So that's really push the will say three and possibly four big players, Ali Baba being one of them as well, coming... where the data has to live in that in that jurisdiction or what have you. But it also has driven fact, better service quality globally to as far as UPTIME, availability. And then back to what he was saying from a Dr Perspective and tail over perspective. So so it's kind of interesting to the evolvement of how policy is driven some of us as well. Yeah, skippy brought up a great point with each region or country can have a different requirement on dat and how it's housed, and we could probably do an advanced show just on that alone right in. So your point, it's not lost. I mean this allows you to be competitive globally with your enterprise globally rather than being restricted within a certain region. Right, essentially helping customers do that and about. Absolutely a lot of our multinational companies that we've worked with have been able to expand globally on all continents. But also now you start, you know, going back to what skip said, you look at California's laws. Essentially it's become its own country when it comes to data. Then having that cloud support that model has been fantastic. Yeah, well, you talked about in and you know you talked about your first experience with s three and then when ect came out. Today we've seen, you know, over the last ten years, tons of services kind of, you know, explode. What are some of the ones outside of compute? Obviously, computing storage are the biggest uses use cases on the clown right, that's the majority of resources. But what are some of those other services that really were game changers that stick out to you guys? I think networking has been a very, very large component to this, particularly for enterprises, because in the past it was tough and maybe more expensive than it should have been to really interface into that hybrid architecture. A lot of the native services on these on all the clock providers have really enabled better connectivity and better experience for the enterprise. And then you look at some of there was an announcement this week ebs multi attached, which is essentially now we have volumes that can be attached to multiple instances, which is a total different transformation than what we've seen with storage before. So you've got block level, you've got object stores, you've got all kinds of it. You can find storage that fits every type of need. A few years ago you kind of had an adopt to what was what was available by the cloud providers, but now the cloud providers are offering you the solutions so that you don't have to transform your architecture as much as you used to. Yeah, I think I specifically in a ws one of the huge Aurora was probably one of the most significant ones. Just in general. That kicked off a lot of this. You know, the database up and in the cloud right. Capabilities Right, if you go back two thousand and twelve, two thousand and thirteen, maybe fourteen, when Aurora was announced. But nevertheless, if you think about having, you know, a relational database being in the cloud right, as opposed to having that high availability, high transactional rate, etc. Having a nonprem the availability to performance, etc. And now you have in the cloud. That was a game changer, I think, and still to this day, as we evolved through right where we out a databases and they're pushing again the other two competitors in that space that to stay up to date with them as well. But customers again gaining all the while. So that's a huge product, or service, if you will, but I think helped. Yeah, you know, the thought that came to mind listen to you guys and just to bring it up maybe just a level right for spen for some of the the late adoptors and people coming in, is the solutions that are available to you guys that are deep in it right are fairly simple. However, those late adoptors, it's not just about computer or storage or databases. It's about how you put these together to create a solution, right. You know, I think back when when we had the early adoptors way in right, they they might have a share point site that they wanted to move to the cloud that, you know, according to the vendors,...

...required so much storage, so much compute, but that was on a future use basis, right. It wasn't on actual usage. And so, depending on your actual scalability or what your actual users required, you can build that for the cloud and save money. That's where we've always seen some issues with customers in the pasture. Let's talk. Let's talk about maybe some of the the new cloud services that maybe five ten years ago, you know, we wouldn't have thought were right around the corner, maybe, whether it's databases or some of the tools like deep learning or for AIML. What do you guys, what are you guys thoughts on that? As I've seen you at reinvent when these are announced and we've talked about it live time, you know. But take a step back five years ago, did you guys envision that some of these things would be out and and with us by two thousand and twenty? I would I would love to say I saw it all coming, but no, absolutely not. But you know, what we've seen is is it's gone from the Lego blocks mentality of I'm going to provide to you the core building blocks so that you can get to the cloud, to the now I actually have a service where you just need to focus on the business outcome and don't need to worry about all the underlying pieces. Sage maker is amazing when it comes to that. On Aws as an example, and it's just going to continue. So I it's getting to the point, though. It used to be very somewhat as a professional in the cloud area, to keep up with everything that's going on as they launched new services, but the expansion of not only the new services but the features within those makes it very hard to keep up, and it's going at such a rate now it's mind blowing. And it's not all three cloud providers are the major cloud providers are there right now and it's no longer they need to worry about compute. That's great. You can't if you want to, because there's going to be a lot of legacy applications out there. But you can take a group of people and say you have a business problem, I'm going to show you how to push data through this pipeline and you're going to get the output that you need. That, to me, is how clad is really beginning. The transformation is accelerating now more than in it at any other time in the cloud history. Yeah, I would dump tell on that, just looking at Lamba, you know those kind of things for service storrytod our architecture again, bringing it out from again, not needing whole staff of code. Pipeline. You know out there Ati Gateway Service. From an AWS perspective. You know, it's a big game changers and I think we're just coming to see a lot of that in order magnitude where it's really business as I finally making that pivot where they're not thinking about infrastructure as a service anymore, even platform as a service. That's really the outcome. At a service is where they are. So it's really exciting, I think. You know. I mean we see it every day, quite frankly, you know. We get pulled into develop scenarios and what have you, you know, and it's really about transforming their business. To help them get to that. will say that the real devots mentality and order operations to go and, you know, again, outcome as a service for a laffo better term. So yeah, let me challenge both of you guys on that. Are you seeing companies actually leverage that? So meaning back, you know, five years ago, one of her favorite sayings was do you want to be the blockbuster, the Netflix? Right, Netflix came out and went from mailing DVDs to this online service. Right, and blockbuster, of course, stuck with brick and mortar and went out of business. And how I would articulate. That's what you guys have said. Is Our companies leverage in this, or do you believe we're going to see some exciting challenges to I wouldn't say status quo, but some of your more entrenched companies, similar to the blockbuster back in the day, where where these young entrepreneurs are going to find a better, faster way to do it with these tools that are out there? Yeah, I think that you can get their better and faster. Definitely now I don't think that's the challenge. In a lot of cases...

...we're dealing with a many hundred year old plus companies out there and they are transforming. They've had a can Tu to transform because they've been here for a hundred years. And what you're seeing is they're starting to create new revenue opportunities, new products, and it's all of these new new technologies. It's areas. They've had these data silos going on back from the beginning of time essentially, and they're unlocking new opportunities and how to use that. It's cool to technology is there to do it, but the bigger risk is are you identifying the business case associated with that and as your organization, your salespeople is that? Are they ready to embrace that change in articulate it back after the customers? And so it's it's getting to the point where the technology is the easier part. It's going to be the business side, it's going to be the more the challenge. Yeah, yeah, same thing, and that's that's the part where, you know, we come in try to help, I shameless sales pitch right, getting people to realize that. So it's the whole ecosystem of the whole thought process. It's challenging, but but time is in our favor because, as you know, I'll say old dinosaurs like me a Jo with the purchasing power and an influence right the younger generation that's coming along, this is all they've ever known. But we need still some infusion of understanding the business outcome, the business impact, the business opportunity. It's cool to play with these things. That are great. They're really if you're a techy and like the dig in and code and all that stuff, you know I can. I can stay at forty dollars in a row and do some of the things, but one of my building and that you know, you take that into the business world and you talk to some of these young folks that are working in this now and the opportunity in front of them and you get to under they get to understand those business opportunities and how to add value. So we're there. I think you see some incremental changes and the future is bright and the skip. I want to bring up a conversation you and I had, sure of you, over a couple cocktails, and it was about outpost. A WS outpost is a new, newer product and it's a physical device that you put on in your own data center. So you're going to actually own a physical device. We were discussing about how there's this generation of people who have never been in a data center that are coming into the industry and how excited they are about seeing some touching a physical server. So I mean, it's that's where we've become and it's just it's a fascinating journey. You know, a lot of this we focused throughout this episode on the history of cloud computer a large majority of that is around aws, right, because they started it. But let's let's kind of talk about and switch gears a little bit about Google and asure and any others that are starting to you know, provide services that you guys believe the enterprises will see where they winning, where they, you know, being competitive, where it does Amazon win and kind of the evolution there and skip maybe just start with your thoughts. Yeah, I think. Well, it's a one, a good one to put on the death I think you know, it's always important to come in, I'll say, platform agnostic when you're looking at things right. Again, from me, from the approach of the outcome, and we have a lot of areas that when we meet with customers and they're talking about whether their club journey is already under way or what they're thinking of, it's really kind of assessing skills, what they're comfortable with, what how they already work right per se, and I think you know a lot of people are comfortable in the old Microsoft framework, as your plays out, still right. They're very intend on how they're building, as you're in, services that get launched without any physical limitations or, you know, governmental limitation that would have you. They're very pragmatic and how they're actually trying to appease to, will say traditional Microsoft customers. The services that they do are very intential from what they experience has been before, including all the way through just engaging from a buying perspective and consumption model. So I think if you look at that from a tried and true perspective, you know there's there's some synergies there, but you still have to...

...focus on the outcome. Technology Wise. You know your sequel, your database stuff. That is Microsoft inherent, as you're does a fantastic job around that and you know. And then you have the advent of arm, the service fabric framework, is really huge from an azure perspective. GCP, on the other hand, I'll let me in pick up the GC P aspect. Be Interested to have this talk is a point of view before I go and paint the water. So, Oh, this out to be good. I think from the spot on with, as you're an aws side, but the Google cloud side I'm starting to see a lot more traction of the enterprise side more than and this is the moment we've been waiting for, because it truly makes three competitors. They're undoubtedly GCP. Google cloud has been around for a while. I use them. Before I actually use to use GCP to low test may a WS infrastructure, which was always a fun game to do. But before I think my mind it was very developer center at very startup centric and they didn't have an enterprise story. But now you're really starting to see how they know how to sell in that area. They're getting there, they're getting better at it, they have a better story and it's becoming more of a natural fit than it has before. So I still think they've got a little bit ways to go, but you know, there's they're all going to be rapidly evolving anyways. But again, having all three of them going neck and neck at each other on features and technology, but more focusing on the business and enterprise side, I think is a great thing for all of us. You know, I'm going to jump in there real quick because I think what you guys said, you know, is very important, especially for the the new users. I mean we see, as you're and ABS still fighting over kind of traditional services and I think in right correct if I'm wrong, but but Google's kind of going after these next Gen services and, as you're in atbs are trying to position themselves there as well. But Google is adapting on the sales front, but from a technology standpoint they can do some really cool next gen type services. And so when you're thinking about rewriting the application. You know, you might still keep your sequel database on ABS or as. Your have that battle. But if you're actually looking at new services that are manageless, you might start looking towards Google becomes more competitive with with ABS and Azures at accurate I think you're absolutely right. Yeah, it's just they have not played into the traditional game like the others have. You know, if you want patching solutions, as you's got great ones, you know, but the kind you can use them across the board. But yeah, it mean computer engines been around for a long time. It was really the first server less model that was out there and I think it was completely underrated. They just didn't know how to necessarily get that out to the enterprise. So, yeah, it was very neat niche as where they were targeting that. So, yeah, table guys, great information today for our listeners. Are on a brief history of the cloud computing and your guys is experienced. You know, listeners out there, how are you going to take this information and use it? Understand really what we're saying is the cloud is evolving faster. It's not becoming more expensive, it's actually becoming cheaper and as you start to evolve your applications out of migration and into the next Gen. You have three different options and really have to look at Your Business and understand that. That's where we are today and that's where we've moved to in this evolution of the cloud. Hey, guys, next week we're going to recap services from last year's reinvent. You know, coming out of there, saw some exciting things. What we want to hear from you up front on the front lines is what our customers asking to use coming out of this reinvent. What are they actually...

...looking at testing or deploying and what are some of the trends you know that maybe change in the last couple of months coming out of reinvent that you know you've noticed, and that's where we'll go next week. Looking forward to it. Yeah, me too, me too really could all right, y'all, thanks for listening in and appreciate it. Send any feedback to cloud crunch at second watchcom. Have a right day. You've been listening to cloud crunch with Jeff aiden, Ian Willoughby and skip Berry. For more information, check out the block second watchcom company block or reach out to second watch on twitter.

In-Stream Audio Search


Search across all episodes within this podcast

Episodes (43)