Transcript of a BriefingsDirect podcast on virtualization strategies and best practices with Bob Meyer, HP's worldwide virtualization lead.
Listen to the podcast. Download the podcast. Find it on iTunes/iPod and Podcast.com. Learn more. Sponsor: Hewlett-Packard.
Dana Gardner: Hi, this is Dana Gardner, principal analyst at Interarbor Solutions, and you're listening to BriefingsDirect. Today, we present a sponsored podcast discussion on the business case and economic rationale for virtualization implementation and best practices.
****Access More HP Resources on Virtualization****
Virtualization has become more attractive to enterprises as they seek to better manage their resources, cut total costs, reduce energy consumption, and improve the agility of their data centers and IT operations. But, virtualization is more than just installing hypervisors. The effects and impacts of virtualization cut across many aspects of IT operations, and the complexity of managing virtualization IT runtime environments can easily slip out of control.
In this podcast, we're going to examine how virtualization can be applied as a larger process and a managed IT undertaking with sufficient tools for governance that allow for rapid, but reasoned, virtualization adoption. We'll show how the proper level of planning and management can more directly assure a substantive economic return on the investments enterprises are making through virtualization.
The goal is to do virtualization right and to be able to scale the use of virtualization in terms of numbers of instances. We also want to extend virtualization from hardware to infrastructure, data, and application support, all with security, control, visibility, and lower risk, and while also helping to make the financial rationale ironclad.
To help provide an in-depth look at how virtualization best practices make for the best economic outcome we're joined by Bob Meyer, the worldwide virtualization lead in Hewlett-Packard’s (HP) Technology Solutions Group (TSG). Welcome to the show, Bob.
Bob Meyer: Thank you very much, Dana.
Gardner: Virtualization is really becoming quite prominent, and we're even seeing instances now where the tough economic climate is accelerating the use and adoption of virtualization. This, of course, presents a number of challenges.
First, could you provide some insight, from HP’s perspective, of how you see virtualization being used in the market now, and how that perhaps has shifted over the past six months or so?
Meyer: When we talk about virtualization -- obviously it’s been around for quite a long time -- it's typically the virtualization of Windows Servers where people start to think about it. For a couple of years now, that’s been the hot value proposition within IT.
The allure there is that when you consider the percentage of budget spent on data center facilities, hardware, and IT operations management, virtualization can have a profound effect on all of these areas.
Moving off the fence
For the last couple of years, people have realized the value in terms of how it can help consolidate servers or how it can help do such things as backup and recovery faster. But, now with the economy taking a turn for the worse, anyone who was on the fence, who wasn’t sure, who didn’t have a lot of experience with it, is now rushing headlong into virtualization. They realize that it touches so many areas of their budget, it just seems to be a logical thing to do in order for them to survive these economic times and come out a leaner, more efficient IT organization.
The change that we see is that previously virtualization was for very targeted use and now it’s gone to virtualization everywhere, for everything -- "How much can I put in and how fast can I put it in."
Gardner: When you move from a tactical orientation to exploit virtualization at this more strategic level, that requires different planning and different methodologies. Tell us what that sort of shift should mean.
Meyer: To be clear, we're not just talking about virtualization of servers. We're talking about virtualizing your infrastructure -- servers, storage, network, and even clients on the desktop. People talk about going headlong into virtualization. It has the potential to change everything within IT and the way IT provides services.
The potential is that you can move your infrastructure around much faster. You can provision a new server in minutes, as opposed to a few days. You can move a virtual machine (VM) from one server to another much faster than you could before.
When you move that into a production environment, if you're talking about it from a services context, a server usually has storage attached to it. It has an IP address, and just because you can move the server around faster doesn’t mean that the IP address gets provisioned any faster or the storage gets attached any faster.
So, when you start moving headlong into virtualization in a production environment, you have to realize that now these are part of services. The business can be affected negatively, if the virtualized infrastructure is managed incompletely or managed outside the norms that you have set up for best practices.
Gardner: I guess it also makes sense that the traditional IT systems-management approaches also need to adjust. If you had standalone stacks, each application with its own underlying platform, physical server, and directly attached data and little bits of middleware for integration, you had a certain setup for managing that. What’s different about managing the virtualized environments, as you are describing them?
Meyer: There are a couple of challenges. First of all, one of the blessings of virtualization is its speed. That’s also a curse in this case, because in traditional IT environments, you set up things like a change advisory board and, if you did a change to a server, if you moved it, if you had to move to a new network segment, or if you had to change storage, you would put it through a change advisory board. There were procedures and processes that people followed and received approvals.
In virtualization, because it’s so easy to move things around and it can be done so quickly, the tendency is for people to say, "Okay, I'm going to ignore that best practice, that governance, and I am going to just do what I do best, which is move the server around quickly and move the storage around." That’s starting to cause all sorts of IT issues.
The other issue is not just the mobility of the infrastructure, but also the visibility of that infrastructure. A lot of the tools that many people have in place today can manage either physical or virtual environments, but not both. What you're heading for when that’s the case is setting up dual management structures. That’s never good for IT. You're just heading for service outages and disruptions when you go in that direction.
Gardner: It sounds like some safeguards are required for managing and allowing automation to do what it does well, but without it spinning out of control and people firing off instances of applications and getting into some significant waste or under-utilization, when in fact that’s what you are trying to avoid.
Shifting the cost
Meyer: Certainly. A lot of what we're seeing is the initial gains of virtualization. People came in and they saw these initial gains in server consolidation. They went from, let’s say, 12 physical boxes down to one physical box with 12 virtual servers. The initial gains get wiped out after a while, and people push the cost from hardware to management, because it becomes harder to manage these dual infrastructures.
Typically, big IT projects get a lot of the visibility. The initial virtualization projects probably get handled with improper procedures. As you come back to day-to-day operations of the virtualized environment, that’s where you start to lose the headway that you gained originally.
That might be from non-optimized infrastructure that is not made to move as fast or to be instrumented as fast as virtualization allows it to be. It could be from management tools that don’t support virtual and physical environments, as we mentioned before. It can even be governance. It can be the will of the IT organization to make sure that they adopt standards that they have in place in this new world of moving and changing environments.
Gardner: For a lot of organizations, with many IT aspects or approaches these days, security and compliance need to be brought into the picture. What does this flexible virtualization capability mean, if you're in a business that has strict compliance and security oversights?
Meyer: Again, it produces its own set of challenges for the reasons similar to what we talked about before. Compliance has many different facets. If you have a service infrastructure that’s in compliance today in a physical environment, it might take days to move that around, and to change the components. People are likely to have much more visibility. That window of change tends to take a lot longer.
With virtualization, because of the speed, the mobility, and the ease of moving things around, things can come out of compliance faster. They could be out of regulatory compliance. They could be out of license compliance, because it’s much easier to spin up new instances of virtual machines and much harder to track them.
So, the same blessing of speed and mobility and ease of instrumentation can take a hit on the compliance and security side as well. It’s harder to keep up with patches. A lot of people do virtual machines through images. They'll create a virtual machine image, and once that image is created, that becomes a static image. You deploy it on one VM and then another and then another. Over time, patches come out, and those patches might not be deployed to that particular image. People are starting to see problems there as well.
Gardner: Just to throw another log on the fire of why this is a complex undertaking, we're probably going to be dealing with hybrid environments, where we have multiple technologies, and multiple types of hypervisors. As you pointed out, the use of virtualization is creeping up beyond servers, through infrastructure storage, and so forth. What’s the hurdle, when it comes to having these mixed and hybrid environments?
Mixed environments are the future
Meyer: That’s a reality that we are going to be dealing with from here on out. Everybody will have a mix of virtual and physical environments. That’s not a technology fad. That’s just a fact. There will be services -- cloud computing, for example -- that will extend that model.
The reality is that the world we live in is both physical and virtual, when it comes to that infrastructure. To have to start looking at it from that perspective, you have to start asking, "Do I have the right solutions in place from an infrastructure perspective, from a management perspective, and from a process perspective to accommodate both environments?"
The danger is having parallel management structures within IT. It does no one any good. If you look at it as a means to an end, which virtualization is, the end of all this is more agile and cost-effective services and more agile and cost-effective use of infrastructure.
Just putting a hypervisor on a machine doesn’t necessarily get you virtualization returns. It allows you to virtualize, but it has to be put on the construct of what you're trying to do. You're trying to provide IT-enabled services for the business at better economies of scale, better agility, and low risk, and that’s the construct that we have to look at.
Gardner: So, if we have a strategic requirement set to prevent some of these blind alleys and pitfalls, then we need to have a strategic process and management overview. This is something that cuts across hardware, software, management, professional conduct and culture, and organization. How do you get started? How do you get to the right level of doing this with that sort of completeness in mind?
Meyer: That’s the problem in a nutshell right there. The way virtualization tends to come in is unique, because it's a revolutionary technology that has the potential to change everything. But, because of the way it comes in, people tend to look at it from a bottom-up perspective. They tend to look at it from, "I have this hypervisor. This hypervisor enables me to do virtual machines. I will manage the hypervisor and the virtual machines, differently than other technologies."
Service-oriented architecture (SOA) and Web services aren't able to creep into an IT environment. They have to come from a top-down perspective. At least somebody has to mandate that they would implement this architecture. So, there's more of a strategy involved.
When we look back at virtualization, the technology is no different than other technologies in the sense that it has to be managed from a strategic perspective. You have to take that top-down look and say, "What does this do for me and for the business?"
At HP, this is where organizations come to us and say, "We have virtualization in our test and development environment, and we are looking to move it into production. What’s the best way to do that?" We come in and assess what they are looking to do, help them roll that up into what’s the bigger picture, what are they trying to get out of this today, and what do they want to get out of this a year from now.
We map out what technologies are in place, how to mix that, how to take the hypervisor environment and make that part of the overall operational management structure, before they move that into the operational environment.
If somebody's already using it and has a number of applications or services they're ready to virtualize, they're already experiencing some of the pain. So, that’s a little bit more prescriptive. Somebody will come in and say, "I'm experiencing this. I'm seeing my management cost rise." Or, "When a service goes down, it’s harder for me to pinpoint where it is, because my infrastructure is more complex."
This is where typically we'll have a spot engagement that then leads to a broader conversation to say, "Let’s fix your pain today, but let’s look at it in the broader context of a service." We have a set of services to do that.
There's a third alternative as well. Sometimes people come to us. They realize the value of virtualization, but they also realize that they don’t have the expertise in house or they don’t have the time to develop that longer-term strategy for themselves. They can also come to HP for outsourcing that virtual and physical environment.
Gardner: It sounds as if the strategic approach to virtualization is similar to what we've encountered in the past, when we've adopted new technologies. We have had to take the same approach of let’s not go just bottom up. Let’s look strategically. Can you offer some examples of how this compares to earlier IT initiatives and how taking that solution approach turned out to be the best cost-benefit approach?
Potential to change everything
Meyer: As an example from an earlier technology perhaps, I always look at client-server computing. When that came out, it had the potential to change everything. If you look at computing today, client-server really did change the way that applications and services were provided.
If you look at the nature of that technology, it required rewriting code and complete architectures. The nature of the technology lent itself to have that strategic view. It was deployed and, over time, a lot of the applications that people were using went to client-server and tier architecture. But, that was because the technology lent itself to that.
Virtualization, in that sense, is not very different. It is a game changer from a top-down perspective. The value you get when you take that top-down perspective is that you have the time to understand that, for example, "I have a set of management tools in place that allow me to monitor my servers, my storage, my network from a service perspective, and they will let me know whether my end users are getting the transaction rates they need on their Web services."
Gardner: Let me just explore that a little bit more. Back when client-server arrived, it wasn’t simply a matter of installing the application on the server and then installing the client on the PCs. Suddenly, there were impacts on the network. Then, there were impacts on the size of the server and capabilities in maintaining simultaneous connections, which required a different approach to the platform.
Then, of course, there was a need for extending this out to branch offices and for wider area networks to be involved. That had a whole other set of issues about performance across the wide area network, the speed of the network, and so on -- a ripple effect. Is that what we're seeing as well with virtualization?
Meyer: We do, absolutely. With the bottom-up approach, people look at it from a hypervisor and a server perspective. But, it really does touch everything that you do, and that everything is not just from a hardware perspective. It not only touches the server itself or the links between the server, the storage, and the network, but it also touches the management infrastructure and the client infrastructure.
So, even though it’s easier to deploy and it can seep in, it touches just about everything. That’s why we keep coming back to this notion of saying that you need to take a strategic look at it, because the more you deploy, the more it will have that ripple effect, as you call it, on all the other systems within IT, and not just a server and hypervisor.
Gardner: Tell us about HP’s history with virtualization. How long has HP been involved with it, and what’s its place and role in the market right now?
Meyer: HP has been doing virtualization for a long time. When most people think of virtualization, they tend to think of hypervisors and they tend to think of it on x86 or Windows servers. That’s really where it has caused this to become popular. But HP has had virtualization in it for quite a while, and we've been doing virtualization on networks for quite a while. So, we are not newcomers to the game.
When it comes to where we play today, there are companies that are experts on the x86 world, and they're providing hypervisors. VMware, Citrix, and Microsoft are really good at what they do. HP doesn’t intend to do that.
What we intend to do is take that hypervisor and make sure that it's part of a well-managed infrastructure, a well-managed service, well-managed desktops, and bringing virtualization into the IT ecosystem, making it part of your day-to-day management fabric.
That’s what we do with hardware that’s optimized out of the box for virtualization. You can wire your hardware once and, as you move your virtual components around, the hardware can take care of the rewiring, the IP network, the IP address, and the storage.
We handle that with IT operations and management offerings that have one solution to heterogeneously manage virtual and physical environments. We do that with client architecture, so that you can extend virtualization onto the desktops, secure the desktops, and take a lot of the cost out of managing them. If you look at what HP is about, it’s taking that hypervisor and actually delivering business value out of a virtual environment.
Gardner: Of course, HP is also in the server hardware business. Does that provide you a benefit in virtualization? Some conventional thinking might be, well gee, why would the hardware people want to increase utilization? Aren’t they in the business of selling more standalone servers?
Meyer: Certainly, we're in the business of selling hardware as well, but the benefit comes in many different areas. Actually, more people today are running virtualization on HP servers than any other platform out there. So, virtualization is an area that allows us to be more creative and more innovative in a server environment.
One of the hottest areas right now in server growth is in blade servers, where you have a bladed enclosure that’s made specifically for virtualization. It allows you to lower the cost of power and cooling, lower the floor space of the data center, and move your virtual components around much faster. Where we might see utilization rates decline in some areas, we're certainly seeing the uptake in others. So, it’s certainly an opportunity for us.
Gardner: So, helping your clients cut the total cost of computing is what’s going to keep you in the hardware business in the long run?
Meyer: That’s exactly right. If you look at the overall benefits, the immediate allure of virtualization is all about the cost and the agility of the service. If you look at it from the bigger picture, if you get virtualization right, and you get it right from a strategic perspective, that’s when you start to feel those gains that we were talking about.
Data centers are very expensive. There's floor space in there. Power and cooling are very expensive. People are talking about that. If we help them get that right and knock the cost out of the infrastructure, the management, the client architectures, and even insourcing or outsourcing, that’s beneficial to everyone.
What are the payoffs?
Gardner: We've talked about how virtualization is a big deal in the market and how it’s being driven by economic factors. We've looked at how a tactical knee-jerk approach can lead to the opposite affect of higher expense and more complexity. We've recognized that taking an experienced, methodological, strategic approach makes a lot of sense.
Now, what is it that we can get, if we do this right? What are the payoffs? Do you have examples of companies you work with, or perhaps within HP itself? I know you guys have done an awful lot in the past several years to refine and improve your IT spend and efficiency. What are the payoffs if you do this right?
Meyer: There are a number of areas. You can look at it in terms of power and cooling. So right off the bat, you can save 50 percent of your power and cooling, if you get this right and get an infrastructure that works together.
From a client-computing perspective, you can save 30 percent off the cost of client computing, off the management of your client endpoints, if you virtualize the infrastructure.
If you look at outsourcing the infrastructure, the returns are manifold there, because you're really taking not just the cost of running it. You're actually leveraging the combined knowledge of thousands and thousands of people who understand how to run the infrastructure from the experience they have of doing multiple outsourcing.
So, we see particular gains in power and cooling, as I mentioned before, and the cost of administration. We'll see significant gains in server-admin ratios. We'll see a threefold increase in the number of servers that people can manage.
If you look across the specific examples, they really do touch a lot of the core areas that people are looking at today -- power and cooling, the cost of maintaining and instrumenting that infrastructure, and the cost of maintaining desktops.
Gardner: Doesn’t this help too, if you have multiple data centers and you're trying to whittle that down to a more efficient, smaller number? Does virtualization have a role in that?
The next generation
Meyer: Absolutely. Actually, throughout the data center, virtualization is one of those key technologies that help you get to that next generation of the consolidated data center. If you just look at from a consolidation standpoint, a couple of years ago, people were happy to be consolidating five servers into one or six servers into one. When you get this right, do it on the right hardware with the right services setup, 32 to 1 is not uncommon -- a 32-to-1 consolidation rate.
If you think about what that equates to, that’s 32 fewer physical servers, less floor space, less power and cooling. So, when you get it right, you go from, "Yes, I can consolidate and I can consolidate it five to one, six to one or 12 to one" to "I'm consolidating, and I am really having a big impact on the business, because I'm consolidating at 24 to 1 or 32 to 1 ratios." That’s really where the payoff starts coming in.
Gardner: I suppose that while you are consolidating, you might as well look at what applications on which platforms are going to be sunset. So, there's a modernization impact. Virtualization helps you move certain apps out to pasture, maybe reusing the logic and the data in the future. What’s the modernization impact that virtualization can provide?
Meyer: Virtualization is absolutely an enabler of that in a number of different ways. Sometimes, when people are modernizing apps, they go to our outsourcing business and say, "I'm modernizing an application and I need some compute capacity. Do you have it?" They can tap into our compute capacity in a virtual way to provide a service, while they're moving, updating, or modernizing an architecture, and the end user doesn’t notice the difference. There's a continuity aspect there, as they provide the application.
There are also the backup and recovery aspects of it. There are a lot of safeguards that come in while you are modernizing applications. In this case, virtualization is an enabler for that. It allows that move to happen. Then, as that application moves onto more up-to-date or more modern architecture, it allows you to quickly scale up or scale down the capacity of that application. Again, the end user experience isn't diminished.
Gardner: So, these days when we are not just dealing with the dollars-and-cents impacts of the economy, we are also looking at dynamic business environments, where there are mergers, acquisitions, bankruptcies, and certain departments being sloughed off, sold, or liquidated. It sounds like the strategic approach to virtualization has a business outcome in that environment too.
Meyer: That’s really where the sort of the flip side of virtualization comes in -- the automation side. Virtualization allows you to quickly spin up capacity and do a series of other things, but automation allows you to do that at scale.
If you have a business that needs to change seasonally, daily, weekly, or at certain times, you need to make much more effective use of that compute capacity. We talk a lot about cost, but it’s automation that makes it cost effective and agile at the same time. It allows you to take a prescribed set of tasks related to virtualization, whether that’s moving a workload, updating a new service, or updating an entire stack and make that happen much faster and at much lower cost, as well.
Gardner: One last area, Bob. I want to get into the benefits of managed virtualization as insurance for the future. You mentioned cloud computing a little earlier. If you do this properly, you start moving toward what we call on-premises or private clouds. You create a fabric of storage, or a fabric of application support, or a fabric of platform infrastructure support. That’s where we get into some of those even larger economic benefits.
This is a vision for many people now, but doing virtualization right seems to me like a precursor to being able to move toward that. You might even be able to start employing SOA more liberally, and then take advantage of external clouds, and there is a whole vision around that. Am I correct in assuming that virtualization is an initial pillar to manage, before you're able to start realizing any of that vision?
Meyer: Certainly. The focus right now is, "How does it save me money?" But, the longer-term benefit, the added benefit, is that, at some point the economy will turn better, as it always does. That will allow you to expand your services and really look at some of the newer ways to offer services. We mentioned cloud computing before. It will be about coming out of this downturn more agile, more adaptable, and more optimized.
No matter where your services are going -- whether you're going to look at cloud computing or enacting SOA now or in the near future -- it has that longer term benefit of saying, "It helps me now, but it really sets me up for success later."
We fundamentally believe, and CIOs have told us a number of times that virtualization will set them up for long-term success. They believe it’s one of those fundamental technologies that will separate their company as winners going into any economic upturn.
Gardner: So, making virtualization a core competency, sooner rather than later, puts you at an advantage across a number of levels, but also over a longer period of time?
Meyer: Yes. Right now everybody is reacting to an economic climate. Those CIOs who are acting with foresight, looking ahead and saying, "Where will this take me," are the ones who are going to be successful as opposed to the people who are just reacting to the current environment and looking to cut and slash. Virtualization has a couple of benefits that allow you to save and optimize, but also sets you up for that -- to boomerang you whenever the economic recovery comes.
Gardner: Well, great. We've been talking with Bob Meyer, the worldwide virtualization lead in HP’s Technology Solutions Group. We've been examining the effects and impacts of virtualization adoption and how to produce the best businesses and financial outcomes from your virtualization initiatives. I want to thank you, Bob, for joining us. It's been a very interesting discussion.
Meyer: Thank you for the opportunity.
****Access More HP Resources on Virtualization****
Gardner: We also want to thank our sponsor, Hewlett-Packard, for supporting this series of podcasts. This is Dana Gardner, principal analyst at Interarbor Solutions. You've been listening to BriefingsDirect. Thanks and come back next time.
Listen to the podcast. Download the podcast. Find it on iTunes/iPod and Podcast.com. Learn more. Sponsor: Hewlett-Packard.
Transcript of BriefingsDirect podcast on virtualization strategies and best practices with Bob Meyer, HP's worldwide virtualization lead. Copyright Interarbor Solutions, LLC, 2005-2009. All rights reserved.