 From theCUBE Studios in Palo Alto in Boston, connecting with thought leaders all around the world, this is a CUBE Conversation. Hi, and welcome to a special CUBE Conversation. I'm Stu Miniman and we're digging into VMware's vSphere 7 announcement. We've had conversations with some of the executives, some of the technical people, but we know that there's no better way to really understand the technology than to talk to some of the practitioners that are using it. Really happy to have joined me for the program. I have Phil Buckley-Meller, who is an infrastructure designer with British Telecom, joining me digitally from across the pond. Phil, thanks so much for joining us. Hi, Stu. All right, so Phil, let's start. Of course, British Telecom, I think most people know what BT is and it's a really sprawling company. Tell us a little bit about your group, your role and what's your mandate. Okay, so my group is called Service Platforms. It's the bit of BT that services all of our multi-millions of our customers. So we have broadband, we have TV, we have mobile, we have DNS and email systems and it's all about our customers. It's not a B2B part of BT, you with me. We specifically focus on those kind of multi-million customers that we've got in those various services. I mean, in particular, my group is for, we do infrastructure, so we really do from data center all the way up to really about boot time or so we'll just pass boot time and the application developers look after that stage and above. Okay, great. We definitely are going to want to dig in and talk about that boundary between the infrastructure teams and the application teams. But let's talk a little bit first. We're talking about VMware. So how long's your organization been doing VMware and tell us what you see with the announcement that VMware is making for BCR7? Sure, well, I mean, we've had a really great relationship with VMware for about 12, 13 years, something like that. And it's an absolutely key part of our infrastructure. It's written throughout BT really in every part of our operations, design, development and the whole ethos of the company is based around a lot of VMware products. And so one of the challenges that we've got right now is application architectures are changing quite significantly at the moment. And as you know, in particular with serverless and with containers and a whole bunch of other things like that, we're very comfortable with our ability to manage VMs and have been for a while. We currently use extensively, we use vSphere, NSXT, VROPS, log insight, network insight and a whole bunch of other VMware constellation applications. And our operations teams know how to use that, they know how to optimize, they know how to capacity plan and troubleshoot. So that's great. And that's been like that for a half a decade at least, we've been really, really confident with our ability to deal with VMware environments. And along came containers and like I say, multi-clouds as well. And what we were struggling with was the inability to have a small pane of glass really on all of that and to use the same people and the same processes to manage a different kind of technology. So we've been working pretty closely with VMware on a number of containerization products for several years. Now I worked really closely with the vSphere integrated containers guys in particular and now with the Pacific guys with really the ideal that when we bring in version seven and the containerization aspects of version seven, we'll be in a position to have that single pane of glass to allow our operations team to really barely differentiate between what's a VM and what's a container. That's really the holy grail, right? So we'll be able to allow our developers to develop our operations team to deploy and to operate and our designers to see the same infrastructure whether that's on-premises, cloud or off-premises and be able to manage the whole piece in that respect. Okay, so Phil, really interesting things you walk through here. You've been using containers in a virtualized environment for a number of years. I want to understand in the organizational piece just a little bit because it sounds great. I manage all the environment, but containers are a little bit different than VMs. If I think back from an application standpoint, it was, let's stick it in a VM. I don't need to change it. And once I spin up a VM, often that's going to sit there for months, if not years, as opposed to, I think about a containerization environment. I really want a pool of resources. I'm going to create and destroy things all the time. So bring us inside that organizational piece. How much will there need to be interaction and more interaction or change in policies between your infrastructure team and your app dev team? Well, yes. I mean, you're absolutely right that the nature and the time scales that we're talking about between VMs and containers is wildly different. As you say, we probably almost certainly have VMs in place now that we're in place in 2018, certainly, I imagine. I haven't really been touched whereas as you say, VMs. A lot of people talk about spinning them all up all the time. There are parts of our architecture that require that. In particular, the very client-facing bursty stuff, it does require spinning up and spinning down pretty quickly. But some of the containers do sit around for weeks, if not months, it really does depend on the development cycle aspects of that. But the hard bit that we've really had was just the visualizing it. There are a number of different products out there that allow you to see the behavior of your containers and understand the resource requirements that they are having at any given moment, allow you to troubleshoot and so on. But they are new products, they're new things that we would have to get used to. And also it seems like there's an awful lot of competing products, quite a Venn diagram in terms of functionality and user abilities to do that. So, again, coming back to being able to manage through vSphere, to be able to have a list of VMs and alongside it is a list of containers and to be able to use policies to define how they behave in terms of their networking, to be able to essentially put our deployments on rails by using, in particular, tag-based policies, means that we can take the onus of security, we can take the onus of performance management and capacity management away from the developers we don't really care about that a lot of the time and they can just get on with their job which is to develop new functionality and help our customers. So, that then means that then we have to be really responsible about defining those policies and making sure that they're adhered to, but again, we know how to do that with VMs through vSphere. So, the fact that we can actually apply that straight away just to a slightly different compute unit, which is really what we're talking about here, is ideal. And then to be able to extend that into multiple clouds as well, because we do use multiple clouds, we're AWS and there's your customers and work between them, is an opportunity that we can't do anything other than be excited about and to take up. Yeah, Phil, I really like how you described really the changing roles that are happening there in your organization need to understand, right, there's things that developers care about, they want to move fast, they want to be able to build new things and there's things that they shouldn't have to worry about and we talk about some of the new world and it's like, oh, can the platform underneath just take care of it? Well, there's some things platforms take care of, there's some things that the software or your team is going to need to understand. So, maybe if you could dig in a little bit, some of those, what are the drivers from your application portfolio? What is the business asking of your organization that's driving this change and being one of those, Tailwind's pushing you towards Kubernetes and the Visture 7 technologies? Well, it all comes down to the customers, right? Our customers want new functionality, they want new integrations, they want new content, they want better stability and better performance and our ability to extend or contract in capacity as needed as well. So, they're the real ultimate challenges that we want to give our customers the best possible experience of our products and services. So, we have to address that really from a development perspective. It's our developers that have the responsibility to design and deploy those. So, we have to, in infrastructure, we have to act as a firm foundation really underneath all of that, that allows them to know that what they spend their time and develop and want to push out to our customers is something that can be trusted, is performance, we understand where the capacity requirements are coming from in the short term and in the long term for that and is secure as well, obviously, is a big aspect to it. So really, we're just providing our developers with the best possible chance of giving our customers what will hopefully make them delighted. Great, Phil, you've mentioned a couple of times that you're using public clouds as well as your VMware farm. Want to make sure, if you can explain a little bit, a couple of things. Number one is when it comes to your team, especially your infrastructure team, how much are they involved with setting up some of the basic pieces or managing things like performance in the public cloud? And secondly, when you look at your applications or some of your cloud, some of your applications, hybrid going between the data center and the public cloud and I haven't talked to too many customers that are doing applications that just live in any cloud and move things around. But maybe if you can clarify those pieces as to what cloud really means to your organization and your applications. Sure, well, I mean, to us, cloud allows us to accelerate development, which is nice because it means that we don't have to do on-premises capacity uplifts for new pieces of functionality or so. We can initially build in the cloud and test in the cloud. But very often applications really make better sense, especially in the TV environment where people watch TV all the time. I mean, yes, there are peak hours and lighter hours of TV watching. Same goes for broadband really. But we generally, we're a, well, more than an eight hour application profile. So what that allows us to do then is to have applications that when it makes sense, we run them inside our organization where we have to run them in our organization for data protection reasons or whatever, then we can do that as well. But where we, say for instance, we have a boxing match on and we're going to be seeing enormous spike in the amount of customers that want to sign up into our order journey for, to allow them to view that and to gain access to that. Well, why would you spend a lot of money on servers just for that level of additional capacity? So we do absolutely have hybrid applications, not necessarily hybrid blocks. We have blocks of sub applications, dozens of them really to support our whole platform. And what you would see is that if you were to look at our full application structure for one of the platforms that I mentioned, that some of the, some of those application blocks have to run inside some can run outside. And what we want to be able to do is to allow our operations team to define that again by policy as to where they run and to have a system that allows us to transparently see where they're running, how they're running and the implications of those decisions. So that we can tune those maybe in the future as well. And that way we best serve our customers. We get to get our customers what they need. All right, great. Phil, final question I have for you. You've been through a few iterations of looking at VMs, containers, public cloud. What advice would you give your peers with the announcement of vSphere 7 and how they can look at things today in 2020 versus what they might have looked at, say a year or two ago? Well, I'll be honest, I was a little bit surprised by vSphere 7. We knew that VMware were working on trying to make containers on the same level both from a management deployment perspective as VMs. I mean, they're called VMware after all, right? We knew that they were looking at that. But I was surprised by just quite how quickly they've managed to almost completely reinvent the application really. It's, you know, if you look at the whole Tanzu stuff and the mission control stuff, I think a lot of people were blown away by just quite how happy VMware were to reinvent themselves from an application perspective, you know, and to really leap forward. And this is, between version six and seven, I've been following these since version three at least. And it's an absolutely revolutionary change in terms of the overall architecture, the aims to what they want to achieve with the application. And, you know, luckily, the nice thing is, is that if you're used to version six, it's not that big a deal. It's really not that big a deal to move forward at all. It's not such a big change to process and training and things like that. But my word, there's an awful lot of work underneath that, underneath the covers. And I'm really excited. And I think all the people in my position should really just take it as an opportunity to revisit, well, revisit what they can achieve with, in particular with vSphere and with In Combination with NSXT, it's quite hard to put into place unless you've seen the slides about it and unless you've seen the product, just how revolutionary the version seven is compared to previous versions, which have kind of evolved for a couple of years. So yeah, I think I'm really excited about it. And I know a lot of my peers or the companies that I speak with quite often are very excited about seven as well. So yeah, I'm really excited about the whole piece. Well, Phil, thank you so much. Absolutely, no doubt this is a huge move for VMware, the entire company and their ecosystem rallying around, help move to the next phase of where application developers and infrastructure need to go. Phil Buckley, joining us from British Telecom. I'm Stu Miniman. Thank you so much for watching theCUBE.