Episode #9: Infrastructure - Why your Company's Servers in the Basement are the Future of Your Co...

Thomas Anglero and Pekko Pääryvärintä discuss why companies should reconsider basement servers instead of moving everything to the cloud. They cover hybrid models, total cost of ownership, and how virtualization and containers let businesses modernize on-premises infrastructure without ripping it all out.
Chapters
Computers now have primary control of critical vehicle functions.
Welcome to the IBM Keep It Simple Technology Podcast, where each episode we discuss technology, but in a simple way. We keep the explanations simple, but what we spend our time on is emphasizing what this technology means to you and how you can best utilize this technology for you and your business. Your host, as always, Thomas Anglero. Welcome to the IBM Keep It Simple Technology Podcast. I'm your host, Thomas Anglero, Director of Innovation here at IBM, and we have another amazing episode. This episode is all about infrastructure, and that is not boring anymore. Today's guest is IBM Finland's country manager for systems. Basically, he is the man for infrastructure, and he's young. He's been promoted up the ladder. Pekko, I'm making you seem really good. He's been promoted up the ladder decades ahead of everybody else. He is the future of infrastructure. Ladies and gentlemen, welcome Pekko. Päivi Rinta. Pekka, are you with us?
Thank you. Yes, Thomas Anglero, and thank you for the kind words. I have had a bit different introduction. As mentioned, I've usually put it the other way. I started very late in the '90s and did technical stuff first. And then ran out of technical skills. So they made me a sales guy and then did sales for quite a few years and ran out of sales skills and they made me a manager. So it's a different angle, but yes, thank you very much.
We love you, Pekka. We love you. Listen, the discussion— this is— I'm so excited to talk to you because one of the elephants in the room today is that cloud computing is getting so much attention, huge amounts of attention. So many business people think that they have to go to the cloud, they must go to the cloud. But in reality, every company has a basement full of servers. They have an infrastructure, an existing infrastructure. And so where I want to start this discussion is why is infrastructure important, though? Today's infrastructure What's the benefit of today's infrastructure? Where is it going? So basically what I'm saying to the people listening out there, why is the equipment, all that capital expenditure you spent in your basement, why is it important to have it, to keep it going? And in what scenario should we go to the cloud? Peko, I want to cover this whole thing because this is something no one ever talks about in meetings. They always talk about the cloud, but there is life in the basement, right? There's life in everything we built. So Pekka, could you just start us off? What is the current situation with all the equipment we have in the basement? Is that stuff junk? Should we just move everything to the cloud? Or is that stuff— what is the future of all that equipment we have? Go for it, Pekka.
Yes, we definitely should talk about cloud. That is something I want to highlight. But public cloud alone is usually not the answer to all of the questions. So, so there's definitely a big play in the hybrid world where you have local infrastructure, whether in your own basement or at your local service provider, and then some of the stuff in, in the public cloud. And some of the key areas we talk with—
Hold on a second. You said public cloud. Let's simplify it for everybody. What do you mean by public cloud? Be specific, please.
Public cloud, like the IBM public cloud, like Amazon, like Microsoft, Google Cloud. Okay, you're talking about—
I just want to keep this simple for everybody. So when you talk about, when you say the word public cloud, we have that picture in our head. Please continue. Sorry, go right ahead.
Yes. And those are important, but there's several reasons why the hybrid model is is in my point of view the way forward. The reasons include regulations, so you need to have data in certain areas or certain countries. The reasons being latency. As far as I know, we haven't invented anything faster than the speed of light, so if you need to have low latency, You really need to have your gear near to your production, for example.
Okay, Pekka, hold on. A moment ago you used the word hybrid cloud. Let's explain to everybody what is hybrid, please.
Hybrid cloud is a combination of your on-premises private cloud and the public cloud hosted by the, by the big players.
And what you were saying a moment ago was that there is advantages to having your own, you're keeping your infrastructure. And one of them you were saying, two of them you're saying, one was regulation, the other one was latency, because nothing's quicker than having actual— the latency to your basement is the quickest versus, uh, pinging an Amazon or an IBM cloud, correct?
That is correct. But it's not only limited to those two elements. There's the TCO, licensing cost, and all those are the reasons behind going to hybrid model.
And TCO is total cost of ownership?
Correct. TCO covering not only the hardware or the cloud infrastructure, but it, it covers the, the software licensing cost, it covers the service support, uh, and, and so forth.
Okay, wait a minute, are you saying that the, the TCO, the total cost ownership of the infrastructure, having my own servers and, uh, having my own servers is actually cheaper than the cloud happening, having all my services up in the cloud? I'm sorry, what did you say?
Not always.
Not always. Okay. Can you just be more specific, please?
When, when there's peak loads and uncertainty of the workloads, then the public cloud is a great place to have your infrastructure or part of it in the public cloud. But if you have a stable workload, not many changes, then most likely the TCO on, on private cloud on-premises is the way forward.
Interesting, interesting. What about, what about you mentioned with peak loads and just in terms of innovation, a lot of these companies they have the infrastructure, they have their investment in the basement, but they don't want to fall behind. They want to be innovative. They want to use some of the newest services and things like that. You being an infrastructure guy, right, when it comes to innovation, how do you see it? Can I innovate from my basement or should I innovate with the cloud? Should I do something hybrid, which is a little bit of both? How do you see going forward with all the newest services? You know, when we were talking offline, you were saying that you've been working with all the acquisitions that IBM have made, all these companies who have the latest technology. Can you give us a— explain how does that work? Innovation in the world of infrastructure versus innovation in the world of cloud.
Yeah. And luckily, I'm from Finland. And back when Nokia mobile phone was a big company, there was a huge amount of engineers at Nokia. They had a bit of troubles with the later days and they released a big number of engineers into the wild in the Finnish forests. And then, so to put into the perspective, I'm meeting these very skilled people daily, and the competition between these new upcoming companies, startups, is heavy. And all of them are thinking new ways, innovative ways to build their services, build their solutions and portfolio. And it's a combination of public cloud, private cloud, I'm working in the infrastructure part mainly, and as you know, IBM stands for International Business Machines, so we still do have quite a few number of different hardware models and solutions in our portfolio, and I see that being a essential part. What we have done is we have really leveraged all the new technologies like container, container world, Linux world in our portfolio. So no matter what kind of technology you want to run, you can select if you run it on on-premise, in your private cloud, or public cloud. It's not the limiting factor. So whatever you want to do, we don't limit you with our offerings where you want to run it. And that's one of the most interesting parts of working with IBM and IBM portfolio.
Okay, so you're saying that if my investment in the basement, that it's, it's still good, right? It's, it's not antiquated. It's still got a future going forward. Um, can you give me, can you give me a feeling for, uh, what industries do you see, um, taking that first step into being more innovative, uh, with their existing infrastructure? Because I think a lot of people— the point I'm getting to is this: a lot of companies aren't doing so much because they're afraid or they say they don't have budget. And let's be realistic, with this whole coronavirus thing going forward, the budgets won't be that big, but innovation will be more important than ever. So in your view, what do you see as the low-hanging fruit going forward with people with the infrastructure who've spent millions and millions in infrastructure and they need to innovate, but their budget's not that big? Can you give us some advice as to where was, what's the first place to look? What's the first thing to do? How do we wrap our heads around this new world that we're in?
Yeah, well, from my perspective, it all comes down to data. So if you are in industry who is built on data, and I really cannot think many industry companies today who have not built their core around data. So I see this, even if you wanted to have an example of some industry, I would say this applies to all industries because most of the future businesses will be digital. And that means it's around data. So you need to be able to handle the data better, more cost efficiently, whether it's due to Corona, whether it's due to competition. There's many reasons you need to do that. But of course, probably the Corona will put pressure on the financial side of the many companies and organizations, and all of those, I believe, still need to look into how they handle data and how to optimize the different data sources. And that's where we can help with different virtualization technologies, where we make the infrastructure platform agnostic, so we don't need to care about whether this runs in public cloud, whether this runs in, in private cloud. We can spread all of this through the virtualization technologies we at IBM have so that the data is available for artificial intelligence or high-performance computing or whatever kind of needs you may have.
When it comes to implementing this data in terms of time period, I'm just thinking that a lot of people out there listening to what you're saying, and you talked about virtualizations, and virtualization been around for some time, and it sounds interesting, but with the, you know, when we all get back to work We're going to be, as I said a moment ago, we're going to need to innovate. How fast can I start innovating with my company, with my— okay, let's do a scenario because I think most people in this scenario, there's two scenarios. One is companies has infrastructure in the basement, nothing in the cloud. Another company is they have infrastructure in the basement and in the cloud. How fast can I get to be more innovative and look at it from both those different groups. How long does it take to do that?
How long? Well, our virtualization capabilities are available in, for example, in AWS. So, click, and you can start using our virtualization for storage, for example. If you have your own hardware in your own premises, I would say from days to weeks to get it up and running. So it's not like it's a year-long project. Of course, it needs to be carefully planned and implemented, but it's not something that you need to wait until your next hardware refresh cycle or or something like that. So, so you can start right now and the immediate savings—
that's, that's fantastic. And what about the complications? I mean, what is this? Does this take special resources, special knowledge? Um, do you have to hire outside companies, or would the typical IT staff of a company be able to do this and manage it and, and help to innovate on that new platform?
Well, that's the one of the fanciest point is that Part of the TCO is the human aspect. You can start with your own IT team or your service provider's IT team. There's not really that much change in the way they would do daily business. Usually it means heaps less work to do because everything is virtualized. But from a skills perspective, most of the guys and girls in the IT team can pick this up really fast.
And let's keep it simple again. And we've used the word virtual many times now. So a moment ago you said virtualized. Can you simplify that for everybody? What do you mean by virtualized?
By virtualizing, we create a software layer between the hardware and the application. With that software layer, we can do different kind of data reduction technologies, so you don't need to buy that much physical hardware. We can do data optimization, so the less frequently used data can be moved to a cheaper hardware or public cloud. So we can do optimization like that. Likewise, in the server side, we can virtualize applications into the containers. So you need less hardware, less CAPEX, and optimize your cost that way.
Got it. You know, I want to touch on a word that is becoming a new buzzword, and a lot of leaders out there haven't heard it, but they're going to hear it more and more when they have their IT meetings. That's the Kubernetes, right? And I just want to give everybody a heads up because I think in 2020, this will be— this word you can hear all the time. And we touched on virtualization, and that's been around for a while, and you've done a wonderful job explaining what that is. But how does— what is this word Kubernetes? What does that mean to my IT infrastructure? How is that going to save me money? Just to take it from a very simple point of view, what is Kubernetes and why is it important to me as a business leader?
Yes, typically applications need operating system to, to, to run and also In the case of Kubernetes, you need that. But the difference between traditional virtualization where you always had the operating system with the application and multiplied that by the number of applications, you created a huge number of wasted space, resources, money for the operating system layer. In the Kubernetes world, you basically remove the copies of the operating system and you make the applications somewhat independent. This not only saves, but it enables you to run those applications wherever. So, for example, with the Red Hat OpenShift that, that we acquired a while ago, you can actually run those container Kubernetes environment in public cloud. You can move those to your on-premises. No matter what cloud provider or on-premise provider you use. And as you know, Linux has been a big— has played a big role within IBM since the early days of Linux. And Kubernetes and the new applications are really based on the open source world., and that's what we have been embracing for quite some time.
So, okay, so let me interrupt you here and let's paint a picture and keep this really simple for everybody and correct me if I'm wrong. So Kubernetes is really the next generation of how we develop on these platforms, the virtualization world. You needed a hardware, you need an operating system and you develop the application and you can have thousands or tens of thousands. As you said, it could be incredibly inefficient. We're using up huge amounts of space, you're buying tons and tons of servers. But some of these services, applications aren't being utilized, but they're just sitting here occupying space. But in the Kubernetes world, it's much more efficient, is lighter, is faster, is portable, is able to work on the cloud, is able to work on your infrastructure. It takes up less space. So you're telling me— did I get that right? Is that right?
First of all, that is exactly what I meant.
Perfect. So that's— so to the business leaders out there, what they should be understanding is that this new world of Kubernetes— now Kubernetes has been around a while, but this new buzz term you're going to be hearing more and more— when you move your— it's not even infrastructure— when your infrastructure is running Kubernetes, you actually could get more out of your existing infrastructure than you can from the world of virtualization. So your IT department, to all your business leaders out there, You spent a fortune on virtualizing everything. I'm sorry. You did a great job. Congratulations. Nothing wrong with that. But this new world of Kubernetes is that you actually can get more life out of your existing expenditures of your stuff, your infrastructure in the basement. Is that correct?
That's correct. So it has evolved into, into new world.
Yeah. And I think it's so important because not enough business leaders hear this simple explanation and understand the difference, and now they do, and why it's important to them and why it benefits them. So is there a cost of moving my infrastructure from a virtualization infrastructure to a Kubernetes-based infrastructure? What is that type of journey? Is that a long journey? Does it take— let's do it timewise. Does that take months to move that application from virtualization to Kubernetes? And then what is the cost to move it from virtualization to Kubernetes?
There is a path, so it is a journey depending on the applications, how they have been developed. There's of course a lot of legacy applications and some of them are easier, some of them are harder. So unfortunately, Thomas, I cannot say you right now that it's going to be an easy path. For everybody, but it's a path that most companies and organizations should start looking at right now and take their planning right now.
And is that— and that's just— is it— when I want— let's go into expense, because every business, every CEO, every leader wants to know how expensive this is going to be. Is this just— this is another development project and a development project per application?
Yeah. Yes, it's, it is the application and the application lifecycle and how those are being cloudified to the new world. So making cloud native.
Yeah, I think this, I think that picture is so important and it's, and it's quite all right. I wasn't trying to put you in a corner that you couldn't answer it. And your answer is 100% correct. Every application has to be looked at individually. And there'll be a certain time frame and cost for each individual one. But to the business leaders out there, what that's saying is— and you're Mr. Infrastructure Man, so I'm trying to back you up on this— and that is that your cost of your infrastructure, that wonderful basement full of all those noisy things with all that air conditioning down there, you had a pump in there, will live on. That is not antiquated. That still has life in it. And that the infrastructure world, the world of hardware and servers, is not dead. Even though the hype is definitely on the cloud computing side, there is still life in the basement. Peko, is— do you have any last words or insights do you want to share with everybody? Because you've been fantastic. You've covered a lot of things. You've done a great job of keeping it simple. Any last words you want to share with us?
Yeah, I would say the infrastructure will enable the financial journey of moving into the new cloud world for the application. So we can definitely help on that path to reduce the TCO for the infrastructure and making the budget available for the journey to cloud. Awesome.
Well said. Pekko Pääryvärintä. My God, your name is killing me, but I love the Finns, so don't beat me up next time I'm in Finland. I thank you, Pekko, for being on the IBM Keep It Simple Technology podcast. To all of you out there, please subscribe to the podcast, share it with your friends, your colleagues. Stay safe, stay well, and we will see you guys on the next episode. Pekko, take care. Thank you very much. Thank you, everybody.
Take care.
Bye-bye.
More from Thomas

IBM Watson AI Lab for Cancer presentation to Ullern HIgh School / Thomas Anglero

Will You Lose Your Job to AI / Oslo Business Forum / Thomas Anglero & Richard Qwest








