Wednesday, November 30, 2011

Big Data Meets Complex Event Processing: AccelOps Delivers a Better Architecture to Attack the Data Center Monitoring and Analytics Problem

Transcript of a BriefingsDirect podcast on how enterprises can benefit from capturing and analyzing systems data to improve IT management in real-time.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: AccelOps.

Connect with AccelOps: Linkedin, Twitter, Facebook, RSS.

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 how new data and analysis approaches are significantly improving IT operations monitoring, as well as providing stronger security. We'll examine how advances in big data analytics and complex events processing (CEP) can come together to provide deep and real-time, pattern-based insight into large-scale IT operations.

AccelOps has developed the technology to correlate events with relevant data across IT systems, so that operators can gain much better insights faster, and then learn as they go to better predict future problems before they emerge. [Disclosure: AccelOps is a sponsor of BriefingsDirect podcasts.]

With us now to explain how these new solutions can drive better IT monitoring and remediation response -- and keep those critical systems performing at their best -- is our guest, Mahesh Kumar, Vice President of Marketing at AccelOps. Welcome to BriefingsDirect, Mahesh.

Mahesh Kumar: Dana, glad to be here.

Gardner: It's always been difficult to gain and maintain comprehensive and accurate analysis of large-scale IT operations, but it seems, Mahesh, that this is getting more difficult. I think there have been some shifts in computing in general in these environments that makes getting a comprehensive view of what’s going on perhaps more difficult than ever. Is that fair in your estimation?

Kumar: Absolutely, Dana. There are several trends that are fundamentally questioning existing and traditional ways of monitoring a data center.

Gardner: Of course we're seeing lots of virtualization. People are getting into higher levels of density, and so forth. How does that impact the issue about monitoring and knowing what’s going on with your systems? How is virtualization a complexity factor?

Kumar: If you look at trends, there are on average about 10 virtual machines (VMs) to a physical server. Predictions are that this is going to increase to about 50 to 1, maybe higher, with advances in hardware and virtualization technologies. So that’s one trend, the increase in density of VMs is a complicating factor for capacity planning, capacity management, performance management, and security.

Corresponding to this is just the sheer number of VMs being added in the enterprise. Analysts estimate that just in the last few years, we have added as many VMs as there were physical machines. In a very short period of time, you have in effect seen a doubling of the size of the IT management problem. So there are a huge number of VMs to manage and that introduces complexity and a lot of data that is created.

Moreover, your workloads are constantly changing. vMotion and DRS are causing changes to happen in hours, minutes, or even seconds, whereas in the past, it would take a week or two for a new server to be introduced, or a server to be moved from one segment of the network to the other.

So change is happening much more quickly and rapidly than ever before. At the very least, you need monitoring and management that can keep pace with today’s rate of change.

Cloud computing

Cloud computing is another big trend. All analyst research and customer feedback suggests that we're moving to a hybrid model, where you have some workloads on a public cloud, some in a private cloud, and some running in a traditional data center. For this, monitoring has to work in a distributed environment, across multiple controlling parties.

Last but certainly not the least, in a hybrid environment, there is absolutely no clear perimeter that you need to defend from a security perspective. Security has to be pervasive.

Given these new realities, it's no longer possible to separate performance monitoring aspects from security monitoring aspects, because of the distributed nature of the problem. You can’t have two different sets of eyes looking at multiple points of presence, from different angles and then try to piece that together.

Those are some of the trends that are causing a fundamental rethink in how IT monitoring and management systems have to be architected.

Gardner: And even as we're seeing complexity ramp-up in these data centers, many organizations are bringing these data centers together and consolidating them. At the same time, we're seeing more spread of IT into remote locations and offices. And we're seeing more use of mobile and distributed activities for data and applications. So we're not only talking about complexity, but we're talking about scale here.

Every office with voice over IP (VoIP) phones needs some servers and network equipment in their office, and those servers and network equipment have to be secured and their up-time guaranteed.



Kumar: And very geographically distributed scale. To give you an example, every office with voice over IP (VoIP) phones needs some servers and network equipment in their office, and those servers and network equipment have to be secured and their up-time guaranteed.

So what was typically thought of as a remote office now has a mini data center, or at least some elements of a data center, in it. You need your monitoring and management systems to have the reach and can easily and flexibly bring those under management and ensure their availability and security.

Gardner: What are some of the ways that you can think about this differently? I know it’s sort of at a vision level, but typically in the past, people thought about a system and then the management of that system. Now, we have to think about clouds and fabrics. We're just using a different vocabulary to describe IT. I suppose we need to have a different vocabulary to describe how we manage and monitor it as well.

Kumar: The basic problem you need to address is one of analysis. Why is that? As we discussed earlier, the scale of systems is really high. The pace of change is very high. The sheer number of configurations that need to be managed is very large. So there's data explosion here.

Since you have a plethora of information coming at you, the challenge is no longer collection of that information. It's how you analyze that information in a holistic manner and provide consumable and actionable data to your business, so that you're able to actually then prevent problems in the future or respond to any issues in real-time or in near real-time.

You need to nail the real-time analytics problem and this has to be the centerpiece of any monitoring or management platform going forward.

Fire hose of data

Gardner: In the past, this fire hose of data was often brought into a repository, perhaps indexed and analyzed, and then over time reports and analysis would be derived from it. That’s the way that all data was managed.

But we really can't take the time to do that, especially when we have to think about real-time management. Is there a fundamental change in how we approach the data that’s coming from IT systems in order to get a better monitoring and analysis capability?

Kumar: The data has to be analyzed in real-time. By real-time I mean in streaming mode before the data hits the disk. You need to be able to analyze it and make decisions. That's actually a very efficient way of analyzing information. Because you avoid a lot of data sync issues and duplicate data, you can react immediately in real time to remediate systems or provide very early warnings in terms of what is going wrong.

The challenges in doing this streaming-mode analysis are scale and speed. The traditional approaches with pure relational databases alone are not equipped to analyze data in this manner. You need new thinking and new approaches to tackle this analysis problem.

Gardner: Also for issues of security, you don't want to find out about security weaknesses by going back and analyzing a bunch of data in a repository. You want to be able to look and find correlations about what's going on, where attacks might be originating, and how that might be affecting different aspects of your infrastructure.

Attackers may hijack an account or gain access to a server, and then over time, stealthily, be able to collect or capture the information that they are after.



People are trying different types of attacks. So this needs to be in real-time as well. It strikes me that if you want to solve security as well as monitoring, that that is also something that has to be in real-time and not something that you go back to every week or month.

Kumar: You might be familiar with advanced persistent threats (APTs). These are attacks where the attacker tries their best to be invisible. These are not the brute-force attacks that we have witnessed in the past. Attackers may hijack an account or gain access to a server, and then over time, stealthily, be able to collect or capture the information that they are after.

These kinds of threats cannot be effectively handled only by looking at data historically, because these are activities that are happening in real-time, and there are very, very weak signals that need to be interpreted, and there is a time element of what else is happening at that time. What seems like disparate sets of activity have to be brought together to be able to provide a level of defense or a defense mechanism against these APTs. This too calls for streaming-mode analysis.

If you notice, for example, someone accessing a server, a database administrator accessing a server for which they have an admin account, it gives you a certain amount of feedback around that activity. But if on the other hand, you learn that a user is accessing a database server for which they don’t have the right level of privileges, it may be a red flag.

You need to be able to connect this red flag that you identify in one instance with the same user trying to do other activity in different kinds of systems. And you need to do that over long periods of time in order to defend yourself against APTs.

Advances in IT

Gardner: So we have the modern data center, we have issues of complexity and virtualization, we have scale, we have data as a deluge, and we need to do something fast in real-time and consistently to learn and relearn and derive correlations.

It turns out that there are some advances in IT over the past several years that have been applied to solve other problems that can be brought to bear here.

This is one of the things that really jumped out at me when I did my initial briefing with AccelOps. You've looked at what's being done with big data and in-memory architectures, and you've also looked at some of the great work that’s been done in services-oriented architecture (SOA) and CEP, and you've put these together in an interesting way.

Let's talk about what the architecture needs to be in order to start doing for IT what we have been doing with retail data or looking at complex events in a financial environment to derive inference into what's going on in the real world. What is the right architecture, now that we need to move to for this higher level of operations and monitoring?

Kumar: Excellent point, Dana. Clearly, based on what we've discussed, there is a big-data angle to this. And, I want to clarify here that big data is not just about volume.

A single configuration setting can have a security implication, a performance implication, an availability implication, and even a capacity implication in some cases.



Doug Laney, a META and a Gartner analyst, probably put it best when he highlighted that big data is about volume, the velocity or the speed with which the data comes in and out, and the variety or the number of different data types and sources that are being indexed and managed. I would add to this a fourth V, which is verdicts, or decisions, that are made. How many decisions are actually impacted or potentially impacted by a slight change in data?

For example, in an IT management paradigm, a single configuration setting can have a security implication, a performance implication, an availability implication, and even a capacity implication in some cases. Just a small change in data has multiple decision points that are affected by it. From our angle, all these different types of criteria affect the big data problem.

When you look at all these different aspects of IT management and how it impacts what essentially presents itself as a big data challenge or a big data problem, that’s an important angle that all IT management and monitoring products need to incorporate in their thinking and in their architectures, because the problem is only going to get worse.

Gardner: Understanding that big data is the issue, and we know what's been done with managing big data in this most comprehensive definition, how can we apply that realistically and practically to IT systems?

It seems to me that you are going to have to do more with the data, cleansing it, discovering it, and making it manageable. Tell me how we can apply the concepts of big data that people have been using in retail and these other applications, and now point that at the IT operations issues and make it applicable and productive.

Couple of approaches

Kumar: I mentioned the analytics ability as central to monitoring systems – big-data analytics to be specific. There are a couple of approaches. Some companies are doing some really interesting work around big-data analysis for IT operations.

They primarily focus on gathering the data, heavily indexing it, and making it available for search, thereby derive analytical results. It allows you to do forensic analysis that you were not easily able to with traditional monitoring systems.

The challenge with that approach is that it swings the pendulum all the way to the other end. Previously we had a very rigid, well-defined relational data-models or data structures, and the index and search approach is much more of a free form. So the pure index-and-search type of an approach is sort of the other end of the spectrum.

What you really need is something that incorporates the best of both worlds and puts that together, and I can explain to you how that can be accomplished with a more modern architecture. To start with, we can't do away with this whole concept of a model or a relationship diagram or entity relationship map. It's really critical for us to maintain that.

I’ll give you an example. When you say that a server is part of a network segment, and a server is connected to a switch in a particular way, it conveys certain meaning. And because of that meaning, you can now automatically apply policies, rules, patterns, and automatically exploit the meaning that you capture purely from that relationship. You can automate a lot of things just by knowing that.

If you stick to a pure index-and-search approach, you basically zero out a lot of this meaning and you lose information in the process.



If you stick to a pure index-and-search approach, you basically zero out a lot of this meaning and you lose information in the process. Then it's the operators who have to handcraft these queries to have to then reestablish this meaning that’s already out there. That can get very, very expensive pretty quickly.

Even at a fairly small scale, you'll find more and more people having to do things, and a pure index and search approach really scales with people, not as much with technology and automation. Index and search certainly adds a positive dimension to traditional IT monitoring tools -- but that alone is not the answer for the future.

Our approach to this big-data analytics problem is to take a hybrid approach. You need a flexible and extensible model that you start with as a foundation, that allows you to then apply meaning on top of that model to all the extended data that you capture and that can be kept in flat files and searched and indexed. You need that hybrid approach in order to get a handle on this problem.

Gardner: I suppose you also have to have your own architecture that can scale. So you're going to concepts like virtual appliances and scaling on-demand vis-à-vis clustering, and taking advantage of in-memory and streaming capabilities to manage this. Tell me why you need to think about the architecture that supports this big data capability in order for it to actually work in practical terms?

Kumar: You start with a fully virtualized architecture, because it allows you not only to scale easily. From a reach standpoint, with a virtualized architecture, you're able to reach into these multiple disparate environments and capture and analyze and bring that information in. So virtualized architecture is absolutely essentially for you to start with.

Auto correlate

Maybe more important is the ability for you to auto-correlate and analyze data, and that analysis has to be distributed analysis. Because whenever you have a big data problem, especially in something like IT management, you're not really sure of the scale of data that you need to analyze and you can never plan for it.

Let me put it another way. There is no server big enough to be able to analyze all of that. You'll always fall short of compute capacity because analysis requirements keep growing. Fundamentally, the architecture has to be one where the analysis is done in a distributed manner. It's easy to add compute capacity by scaling horizontally. Your architecture fits how computing models are evolving over the long run. So there are a lot of synergies to be exploited here by having a distributed analytics framework.

Think of it as applying a MapReduce type of algorithm to IT management problems, so that you can do distributed analysis, and the analysis is highly granular or specific. In IT management problems, it's always about the specificity with which you analyze and detect a problem that makes all the difference between whether that product or the solution is useful for a customer or not.

Gardner: In order for us to meet our requirements around scale and speed, we really have to think about the support underneath these capabilities in a new way. It seems like, in a sense, architecture is destiny when it comes to the support and monitoring for these large volumes in this velocity of data.

Let's look at the other part of this. We talked about the big data, but in order for the solution to work, we're looking at CEP capabilities in an engine that can take that data and then work with it and analyze it for these events and these programmable events and looking for certain patterns.

A major advantage of distributed analytics is that you're freed from the scale-versus-richness trade-off, from the limits on the type of events you can process.



Now that we understand the architecture and why it's important, tell me why this engine brings you to a higher level and differentiates you in the field around the monitoring.

Kumar: A major advantage of distributed analytics is that you're freed from the scale-versus-richness trade-off, from the limits on the type of events you can process. If I wanted to do more complex events and process more complex events, it's a lot easier to add compute capacity by just simply adding VMs and scaling horizontally. That’s a big aspect of automating deep forensic analysis into the data that you're receiving.

I want to add a little bit more about the richness of CEP. It's not just around capturing data and massaging it or looking at it from different angles and events. When we say CEP, we mean it is advanced to the point where it starts to capture how people would actually rationalize and analyze a problem.

For example, the ability for people in a simple visual snapshot to connect three different data points or three events together and say that they're interrelated and they point to a specific problem.

The only way you can automate your monitoring systems end-to-end and get more of the human element out of it is when your CEP system is able to capture those nuances that people in the NOC and SOC would normally use to rationalize when they look at events. You not only look at a stream of events, you ask further questions and then determine the remedy.

No hard limits

To do this, you should have a rich data set to analyze, i.e. there shouldn’t be any hard limits placed on what data can participate in the analysis and you should have the flexibility to easily add new data sources or types of data. So it's very important for the architecture to be able to not only event on data that are is stored in in traditional models or well-defined relational models, but also event against data that’s typically serialized and indexed in flat file databases.

This hybrid approach basically breaks the logjam in terms of creating these systems that are very smart and that could substitute for people in terms of how they think and how they react to events that are manifested in the NOC. You are not bound to data in an inflexible vendor defined model. You can also bring in the more free-form data into the analytics domain and do deep and specific analysis with it.

Cloud and virtualization are also making this possible. Although they’ve introduced more complexity due to change frequency, distributed workloads etc., they’ve also introduced some structure into IT environments. An example here is the use of converged infrastructure (Cisco UCS, HP Blade Matrix) to build private-cloud environments. At least at the infrastructure level it introduces some order and predictability.

Gardner: All right, Mahesh, we've talked about the problem in the market, we have talked about high-level look at the solution and why you need to do things differently, and why having the right architecture to support that is important, but let's get into the results.

If you do this properly, if you leverage and exploit these newer methods in IT -- like big data, analytics, CEP, virtual appliances and clustered instances of workload and support, and when you apply all those to this problem about the fire hose of data coming out of IT systems, a comprehensive look at IT in this fashion -- what do you get? What's the payoff if you do this properly?

Their needs are really around managing security, performance and configurations. These are three interconnected metrics in a virtualized cloud environment.



Kumar: I want to answer this question from a customer standpoint. It is no surprise that our customers don’t come to us saying we have a big data problem, help us solve a big data problem, or we have a complex event problem.

Their needs are really around managing security, performance and configurations. These are three interconnected metrics in a virtualized cloud environment. You can't separate one from the other. And customers say they are so interconnected that they want these managed on a common platform. So they're really coming at it from a business-level or outcome-focused perspective.

What AccelOps does under the covers, is apply techniques such as big-data analysis, complex driven processing, etc., to then solve those problems for the customer. That is the key payoff -- that customer’s key concerns that I just mentioned are addressed in a unified and scalable manner.

An important factor for customer productivity and adoption is the product user-interface. It is not of much use if a product leverages these advanced techniques but makes the user interface complicated -- you end up with the same result as before. So we’ve designed a UI that’s very easy to use, requires one or two clicks to get the information you need; a UI-driven ability to compose rich events and event patterns. Our customers find this very valuable, as they do not need super-specialized skills to work with our product.

Gardner: What's important to think about when we mention your customers is not just applying this value to an enterprise environment. Increasingly the cloud, with the virtualization, the importance of managing performance to very high standards, these are also impacting the cloud providers, managed service providers (MSPs), and software-as-a-service (SaaS) providers.

Up and running

T
his sounds like an architecture, an approach and a solution that's going to really benefit them, because their bread and butter is about keeping all of the systems up and running and making sure that all their service level agreements (SLAs) and contracts are being managed and adhered to.

Just to be clear, we're talking about an approach for a fairly large cross-section of the modern computing world -- enterprises and many different stripes of what we consider as service providers.

Kumar: Service providers are a very significant market segment for us and they are some of our largest customers. The reason they like the architecture that we have, very clearly, is that it's scalable. They know that the architecture scales as their business scales.

They also know that they get both the performance management and the security management aspects in a single platform. They're actually able to differentiate their customer offerings compared to other MSPs that may not have both, because security becomes really critical.

For anyone wanting to outsource to an MSP, the first question or one of the first questions that they are going to ask, in addition to the SLAs, are how are you going to ensure security? So to have both of those options is absolutely critical.

Subscription based licensing, which we offer in addition to perpetual licensing, also fits well with the CSP/MSP business model.



The third piece really is the fact that our architecture is multi-tenant from day one. We're able to bring customers on board with a one-touch mechanism, where they can bring the customer online, apply the right types of policies, whether it's SLA policies or security polices, automatically in our product and completely segment the data from one customer to the other.

All of that capability was built into our products from day one. So we didn’t have to retrofit any of that. That’s something our cloud-service providers and managed service provider customers find very appealing in terms of adopting AccelOps products.

Subscription based licensing, which we offer in addition to perpetual licensing, also fits well with the CSP/MSP business model.

Gardner: All right. Let's introduce your products in a little bit more detail. We understand you have created a platform, an architecture, for doing these issues or solving these issues for these very intense types of environments, for these large customers, enterprises, and service providers. Tell us a little bit about your portfolio.

Key metrics

Kumar: What we've built is a platform that monitors data center performance, security, and configurations. The three key interconnected metrics in virtualized cloud environments. Most of our customers really want that combined and integrated platform. Some of them might choose to start with addressing security, but they soon bring in the performance management aspects into it also. And vice versa.

And we take a holistic cross-domain perspective -- we span server, storage, network, virtualization and applications.

What we've really built is a common consistent platform that addresses these problems of performance, security, and configurations, in a holistic manner and that’s the main thing that our customers buy from us today.

Gardner: It sounds as if we're doing business intelligence for IT. We really are getting to the point where we can have precise dashboards, and we are not just making inferences and guesses. We're not just doing Boolean searches on old or even faulty data.

We're really looking at the true data, the true picture in real-time, and therefore starting to do the analysis that I think can start driving productivity to even newer heights than we have been accustomed to. So is that the vision, business intelligence (BI) for IT?

As you add the number of VMs or devices, you simply cannot scale the management cost, in a linear fashion. You want to have continuously reducing management cost for every new VM added or new device introduced.



Kumar: I guess you could say that. To break it down, from an IT management and monitoring standpoint, it is on an ongoing basis to continuously reducing the per capita management costs. As you add the number of VMs or devices, you simply cannot scale the management cost, in a linear fashion. You want to have continuously reducing management cost for every new VM added or new device introduced.

The way you do that is obviously through automation and through a self-learning process, whereby as you continue to learn more and more about the behavior of your applications and infrastructure, you're able to start to easily codify more and more of those patterns and rules in the system, thereby taking sort of the human element out of it bit by bit.

What we have as a product and a platform is the ability for you to increase the return on investment (ROI) on the platform as you continue to use that platform day-to-day. You add more information and enrich the platform with more rules, more patterns, and complex events that you can detect and potentially take automated actions on in the future.

So we create a virtuous cycle, with our product returning higher and higher return on your investment with time. Whereas, in traditional products, scale and longevity have the opposite effect.

So that’s really our vision. How do you reduce the per capita management cost as the scale of the enterprises start to increase, and how do you increase more automation as one of the elements of reducing the management cost within IT?

Gardner: You have given people a path to start in on this, sort of a crawl-walk-run approach. Tell me how that works. I believe you have a trial download, an opportunity for people to try this out for free.

Free trial download

Kumar: Most of our customers start off with the free trial download. It’s a very simple process. Visit www.accelops.com/download and download a virtual appliance trial that you can install in your data center within your firewall very quickly and easily.

Getting started with the AccelOps product is pretty simple. You fire up the product and enter the credentials needed to access the devices to be monitored. We do most of it agentlessly, and so you just enter the credentials, the range that you want to discover and monitor, and that’s it. You get started that way and you hit Go.

The product then uses this information to determine what’s in the environment. It automatically establishes relationships between them, automatically applies the rules and policies that come out of the box with the product, and some basic thresholds that are already in the product that you can actually start measuring the results. Within a few hours of getting started, you'll have measurable results and trends and graphs and charts to look at and gain benefits from it.

That’s a very simple process, and I encourage all our listeners and readers to download our free trial software and try AccelOps.

Gardner: I also have to imagine that your comments a few moments ago about not being able to continue on the same trajectory when it comes to management is only going to accelerate the need to automate and find the intelligent rather than the hard or laborious way to solve this when we go to things like cloud and increased mobility of workers and distributed computing.

It’s about automation and distributed analytics and about getting very specific with the information that you have, so that you can make absolutely more predictable, 99.9 percent correct of decisions and do that in an automated manner.



So the trends are really in your favor. It seems that as we move toward cloud and mobile that at some point or another organizations will hit the wall and look for the automation alternative.

Kumar: It’s about automation and distributed analytics and about getting very specific with the information that you have, so that you can make absolutely more predictable, 99.9 percent correct of decisions and do that in an automated manner. The only way you can do that is if you have a platform that’s rich enough and scalable and that allows you to then reach that ultimate goal of automating most of the management of these diverse and disparate environments.

That’s something that's sorely lacking in products today. As you said, it's all brute-force today. What we have built is a very elegant, easy-to-use way of managing your IT problems, whether it’s from a security standpoint, performance management standpoint, or configuration standpoint, in a single integrated platform. That's extremely appealing for our customers, both enterprise and cloud-service providers.

I also want to take this opportunity to encourage those of your listening or reading this podcast to come meet our team at the 2011 Gartner Data Center Conference, Dec. 5-9, at Booth 49 and learn more. AccelOps is a silver sponsor of the conference.

Gardner: I am afraid we will have to leave it there. You've been listening to a sponsored BriefingsDirect podcast. We've been talking about how new data and analysis approaches from AccelOps are attaining significantly improved IT operations monitoring as well as stronger security.

I'd like to thank our guest, Mahesh Kumar, Vice President of Marketing at AccelOps. Thank so much, Mahesh.

Kumar: Thank you, Dana.

Gardner: This is Dana Gardner, Principal Analyst at Interarbor Solutions. Thanks again for listening and come back next time.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: AccelOps.

Connect with AccelOps: Linkedin, Twitter, Facebook, RSS.

Transcript of a BriefingsDirect podcast on how enterprises can benefit from capturing and analyzing systems data to improve IT management in real-time. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in:

Tuesday, November 29, 2011

HP Discover Case Study: Vodafone Ireland IT Group Sees Huge ROI By Emphasizing Business Service Delivery

Transcript of a BriefingsDirect podcast in conjunction with HP Discover 2011 in Vienna on how a major telecom provider has improved service to customers by shifting from a technology emphasis to business service delivery.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: HP.

Dana Gardner: Hello, and welcome to a special BriefingsDirect podcast series coming to you in conjunction with the HP Discover 2011 Conference in Vienna.

We’re here in the week of Nov. 28, to explore some major case studies from some of Europe’s leading enterprises. We'll see how a series of innovative solutions and an IT transformation approach to better support business goals is benefiting these companies, their internal users, and their global customers.

I'm Dana Gardner, Principal Analyst at Interarbor Solutions and I'll be your host throughout this series of HP-sponsored Discover live discussions.

Out next customer case study interview highlights how a shift from a technology emphasis to a business services delivery emphasis has created significant improvements for a large telecommunications provider, Vodafone.

To learn more, we’re here with Shane Gaffney, Head of IT operations for Vodafone Ireland in Dublin. Welcome to the show, Shane.

Shane Gaffney: Thank you, Dana.

Gardner: Tell me what was the challenge that you faced when you decided to switch from a focus on technology purely to one more of a business user mentality? Why did you think you needed to do that?

Gaffney: Back in summer of 2010, when we looked at the business perception of the quality of service received from IT, the confidence was lower than we’d like in terms of predictable and optimal service quality being provided.

There was a lack of transparency. Business owners didn’t fully understand what quality was being received and they didn’t have simple meaningful language that they were receiving from IT operations in terms of understanding service quality: good, bad, or indifferent.

Within IT operations, as a function, we also had our own challenges. We were struggling to control our services. We were under the usual pressure that many of our counterparts face in terms of having to do more with less, and downward pressure on cost and headcount. We were growing a dynamic complex IT estate, plus customers are naturally becoming ever more discerning in terms of their expectations of IT.

Transactional and reactive


As a large multinational, we specified a fragmented organization, some services being provided centrally vs. locally, some offshore vs. onshore. With that, we had a number of challenges and the type of relationship we had with our customers was transactional and reactive in nature.

So with that backdrop, we knew we needed to take some radical steps to really drive our business forward.

Gardner: And before we learn more about that, Shane, tell me a little bit about Vodafone Ireland? Tell us the extent of your services and your reach there?

Gaffney: Vodafone is Ireland’s leading telecommunications operator. We have in excess of 2.4 million subscribers, about 1,300 employees in a mixture of on-premise and cloud operations. I mentioned the complex and dynamic IT estate that we manage. To put a bit of color around that, we’ve got 230 applications, about 2,500 infrastructure nodes that we manage either directly or indirectly -- with substantial growth in traffic, particularly the exponential growth in the telecom data market.

Gardner: When you decided to change your emphasis to try to provide more of that business confidence, the services orientation, clearly just using technology to do that probably wasn't going to be sufficient. There are issues around people, process, culture, and so forth. How do you, at a philosophical level, bridge the continuum among and between technology and these other softer issues like culture?

Gaffney: That's a really important point. The first thing we did, Dana, was engage quite heavily with all of our business colleagues to define a service model. In essence what we were looking at there was having our business unit owners define what services were important to them at multiple levels down to the service transactions, and defining the attributes of each of those services that make them successful or not.

We essentially looked to align our people, revamp our processes, and look at our end-to-end tool strategy, all based around that service model.



Once we had a very clear picture of what that looked like across all business functions, we used that as our starting point to be able to measure success through the customer eyes.

That's the focus and continues to be the core driver behind everything else we do in IT operations. We essentially looked to align our people, revamp our processes, and look at our end-to-end tool strategy, all based around that service model.

The service model has enforced a genuine service orientation and customer centricity that’s driven through all activities and behaviors, including the culture within the IT ops group in how we service customers. It’s really incorporating those commercial and business drivers at the heart of how we work.

Gardner: Shane, I've heard from other companies that another important aspect of moving to the shift on services delivery is to gain more awareness of the IT products in total. It involves, I suppose, gaining insight and then analysis, not at the point-by-point basis in the products themselves, but at that higher abstraction of how the users themselves view these services? Has that been important for you as well?

Helicopter view

Gaffney: We’ve taken the service view at a number of levels. Essentially, the service model is defined at a helicopter view, which is really what’s important to our respective customers. And we’ve drilled down into a number of customer or service-oriented views of their services, as well as mapping in, distilling, and simplifying the underlying complexities and event volumes within our IT estate.

Gardner: In order to get that helicopter view and abstract things in terms of a process level, what have you done in terms of gaining insight? What has become important for you to be able to do that?

Gaffney: There are a number of things we’ve considered there. Without having a consolidated or rationalized suite of tools, we found previously that it's very difficult to get control of our services through the various tiers. By introducing the HP Application Performance Management tools portfolio, there are a number of modules therein that have allowed us to achieve the various goals that we’ve set to achieve the desired control.

Gardner: Before we go into any detail on products and approaches, let’s pause and step back. What does this get for you -- if you do it right? What is it that you've been able to attain by shifting your emphasis to the business services level and employing some new approaches in culture? What did you get? What’s the payoff?

Gaffney: First of all for IT, we build confidence within the team in terms of having a better handle on the quality of service that we’re offering. Having that commercial awareness really does drive the team forward. It means that we’re able to engage with our customers in a much more meaningful way to create genuine value-add, and move away from routine transactional activity, to helping our customers to innovate and drive business forward.

Without having a consolidated or rationalized suite of tools, we found previously that it's very difficult to get control of our services through the various tiers.



We’ve certainly enjoyed those type of benefits through our transformation journey by automating a lot of the more core routine and repeatable activity, facilitating focus on our relationship with our customers in terms of understanding their needs and helping them to evolve the business.

Gardner: Have you done any surveys or presented key performance indicators (KPIs) against some of this? Is it still early? How might we look to some more concrete results, if you're able to provide that?

Gaffney: In terms of how we measure success, Dana, we try to take a 360 view of our service quality. So we have a comprehensive suite of KPIs at the technology layer. We also do likewise in terms of our service management and establishing KPIs and service level agreements (SLAs) at the service layer. We've then taken a look at what quality looks like in terms of customer experience and perception, seeking to correlate metrics between these perspectives.

As an example, we routinely and rigorously measure our customer net promoter score, which essentially assesses whether the customers, based on their experience, would recommend our products and services to others.

To give a flavor of the type of KPI improvements at an operational level that we’ve seen improve over the last year, we measure "customer loss hours," which is effectively due to any diminished performance in, or availability of, our services. We measure the impact to the end customer in terms of the adverse impact they would suffer.

Reduction in lost hours

We’ve seen a 66 percent reduction in customer lost hours year on year from last summer to this. We’ve also seen a 75 percent reaction in mean time to repair or average service restoration time.

Another statistic I'd call out briefly is that at the start of this process, we were identifying root cause for incidents that were occurring in about 40-50 percent of cases on average. We’re now tracking consistently between 90-100 percent in those cases and have thereby been able to better understand, through our capabilities and tools, what’s going on in the department and what’s causing issues. We consequently have a much better chance of avoiding repetition in those issues impacting customers.

At a customer satisfaction level, we’ve seen similar improvements that correlate with the improved operational KPIs. From all angles, we’ve thankfully enjoyed very substantial improvements. If we look at this from a financial point of view, we’ve realized a return on investment (ROI) of 300 percent in year one and, looking solely at the cost to fix and the cost of failure in terms of not offering optimal service quality, we’ve been able to realize cost savings in the region of €1.2 million OPEX through this journey.

Gardner: Let me just dig into that ROI. That’s pretty amazing, 300 percent ROI in one year. And what was that investment in? Was that in products, services, consulting, how did you measure it?

At a customer satisfaction level, we’ve seen similar improvements that correlate with the improved operational KPIs.



Gaffney: Yes, the ROI is in terms of the expenditure that would have related primarily to our investment in the HP product portfolio over the last year as well as a smaller number of ancillary solutions.

The payback in terms of the benefits realized from financial perspective that relate to the cost savings associated with having fewer issues and in the event where we have issues, the ability to detect those faster and spend less labor investigating and resorting issues, because the tools, in effect, are doing a lot of that legwork and much of the intelligence is built in to that product portfolio.

Gardner: I suppose this would be a good time to step back and take a look at what you actually do have in place. What specifically does that portfolio consist of for you there at Vodafone Ireland?

Gaffney: We have a number of modules in HP's APM portfolio that I'll talk about briefly. In terms of looking to get a much broader and richer understanding of our end-user experience which we lacked previously, we’ve deployed HP’s Business Process Monitors (BPMs) to effectively emulate the end-user experience from various locations nationwide. That provides us with a consistent measure and baseline of how users experience our services.

We’ve deployed HP Real User Monitoring (RUM), which gives us a comprehensive micro and macro view of the actual customer experience to complement those synthetic transactions that mimic user behavior. Those two views combined provide a rich cocktail for understanding at a service level what our customers are experiencing.

Events correlation

We then looked at events correlation. We were one of the first commercial customers to adopt HP’s BSM version 9.1 deployment, which gives us a single pane of glass into our full service portfolio and the related IT infrastructure.

Looking a little bit more closely at BSM, we've used HP’s Discovery and Dependency Mapping Advanced (DDMa) to build out our service model, i.e. effectively mapping our configuration items throughout the estate, back up to that top-down service view. DDMa effectively acts as an inventory tool that granularly links the estate to service. We’ve aligned the DDMa deployment with our service model which, as I mentioned earlier, is integral to our transformation journey.

Beyond that, we’ve looked at HP’s Operations Manager i (OMI) capability, which we use to correlate our application performance and our system events with our business services. This allows our operators to reduce a lot of the noisy events by distilling those high-volume events into unique actionable events. This allows operators to focus instead on services that may be impacted or need attention and, of course, our customers and our business.

We’ve gone farther and looked at ArcSight Logger, software which we’ve deployed to a single location that collects logged files throughout our estate. This allows us to quickly and easily search across all logged files for abnormalities that might be related to a particular issue.

By integrating ArcSight Logger with OMI -- and I believe we’re one of the first HP customers to do this -- we’ve enriched operator views with security information as well as the hardware, OS, and application layer events. That gives us a composite view of what’s happening with our services through multiple lenses, holistically across our technology landscape and products and services portfolio.

A year ago, we were to a degree reactive in terms of how we provided service. At this point, we’re proactive in how we manage services.



Additionally, we’ve used HP’s Operations Orchestration to automate many of our routine procedures and, picking up on the ROI, this has allowed us to free up operators’ time to focus on value-add and effectively to do more with less. That's been quite a powerful module for us, and we’ve further work to exploit that capability.

The last point to call out in terms of the HP portfolio is we’re one of the early trialists of HP’s Service Health Analyzer. A year ago, we were to a degree reactive in terms of how we provided service. At this point, we’re proactive in how we manage services.

Service Health Analyzer will allow us to move to the next level of our evolution, moving toward predictive service quality. I prefer to call the Service Health Analyzer our “crystal ball,” because that’s essentially what we’re looking at. It’s taking trends that are occurring with the services of transaction, and predicting what's likely to happen next and what may be in jeopardy of breaking down the line, so you can take early intervention and remedial action before there’s any material impact on customers.

We’re quite excited about seeing where we can go there. One of the sub-modules of Service Health Analyzer is Service Health Reporter, and that’s a tool that we expect to act as our primary capacity planning capability across a full IT estate going forward.

Throughout our implementation, partnership was a key ingredient to success. Vodafone had the business vision and appetite to evolve. HP provided the thought leadership and guidance. And, Perform IT, HP's partner, brought hands-on implementation and tuning expertise into the mix.

Gardner: That’s very impressive. You’re certainly juggling a lot of balls and keeping them in the air. One of the things that I've seen in the market when it comes to gaining this sort of pane of glass view into operations is they’re starting to share that as sort of a dashboard, or a graphical representation as a scorecard perhaps we could refer to it, with more of the business leadership.

Have you been able to take some of the analysis, and insights and then not just use that in the context of the IT operations, but provide it back to business, so it would help them manage their strategy and operational decision making?

Full transparency

Gaffney: Absolutely. One of our core principles throughout this journey has been to offer full transparency to our customers in terms of the services they receive and enjoy from us. On one hand, we provide the BSM console to all of our customers to allow them to have a view of exactly what the IT teams see, but with a service orientation.

We’re actually going a step further and we’re building out a cloud-based service portal that takes a rich feed in from the full BSM portfolio, including the modules that I've called out earlier. It also takes feeds in from a remedy system, in order to get the view of core processes such as incident management, problem management, change management.

Bringing all of that information together gives customers a comprehensive view of the services they receive from IT operations. That's our aim -- to provide customers with everything they need at their fingertips.

It's essentially providing simple and meaningful information with customized views and dynamic drill-down capabilities, so customers can look at a very high level of how the services are performing, or really drill into the detail, should they so desire. The portal, we believe, is likely to act as a powerful business enabler. Ultimately, we believe there's opportunity to commercialize or productize this capability down the line.

The portal, we believe, is likely to act as a powerful business enabler. Ultimately, we believe there's opportunity to commercialize or productize this capability down the line.



Gardner: We’re about out of time, but Shane, now that you've gone through quite a bit of this, and as an early adopter, I wonder if you could share some 20-20 hindsight for those users around the world who are examining some of the products and services available, thinking about culture, re-emphasizing the business process issues, rather than just pure technology issues. What would you tell them as advice when they get started? Any recommendations now that you've been through this yourself?

Gaffney: For customers embarking on this type of transformation initiative, first off, I would suggest: engage with your customers. Speak with your customers to deeply understand their services, and let them define what success looks like.

Look to promote quick wins and win-wins. Look at what works for the IT community and what works for the customer. Both are equally important. Buy-in is required, and people across those functions all need to understand what success looks like, and believe in it.

I would recommend taking a holistic approach from a couple of angles. Don’t just look at your people, technology, or processes, but look at those collectively, because they need to work in harmony to hit the service quality sweet spot. Holistically, it's important to prepare your strategy, but look top down from the customer view down into your IT estate and vice versa, mapping all configuration items back into those top level services.

Rationalize and automate

Rationalize and automate wherever possible. We had a suite of over two dozen tools, acting as a cumbersome patchwork solution for operators. We’ve vastly rationalized those tools into a much more manageable single console that the teams use now.

We’ve automated all resource-intensive and transactional activities wherever possible, which again frees up time to allow engineers to focus on the business relationship.

I’d also recommend the people incrementally build on success. We started out with modest budget, but by targeting early wins through that investment, and by building subsequent business cases, particularly with the service model, we were easily able to get the buy-in from stakeholders, because the story was compelling, based on the commercial advantages and the broader business benefits that were accrued from the earlier investment.

Lastly, for IT teams I would strongly suggest that you look to establish a dedicated surveillance capability, whether that’s round the clock or whatever is appropriate for your business model. Moving from a traditional support model to this type of service-oriented view, the key to success is having people managing the eyes and ears across your services at all times. It really does pay back in spades.

We’ve automated all resource-intensive and transactional activities wherever possible, which again frees up time to allow people focus on the business relationship.



Gardner: Excellent. A big thank you to you Shane Gaffney, Head of IT Operations at Vodafone Ireland. This has been a great story, and thank you for sharing it on how a shift from technology emphasis to a business services delivery emphasis has created some significant improvements and has set the stage for yet greater business productivity from IT.

I also want to thank our audience for joining us for this special BriefingsDirect podcast coming to you in conjunction with the HP Discover 2011 Conference in Vienna. I hope you have a great show. I appreciate your time, Shane.

Gaffney: Thank you, Dana.

Gardner: This is Dana Gardner, Principal Analyst at Interarbor Solutions, your host for this series of HP-sponsored Discover live discussions. Thanks again for listening and come back next time.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: HP.

Transcript of a BriefingsDirect podcast in conjunction with HP Discover 2011 in Vienna on how a major telecom provider has improved service to customers by shifting from a technology emphasis to business service delivery. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in:

Tuesday, November 15, 2011

Germany's Largest Travel Agency Starts a Virtual Journey to Get Branch Office IT Under Control

Transcript of a sponsored podcast discussion from VMworld 2011 in Copenhagen on how DER Deutsches Reisebüro virtualized 2,300 desktops to centralize administration.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: VMware.

Dana Gardner: Hello, and welcome to a special BriefingsDirect podcast series coming to you from the VMworld 2011 Conference in Copenhagen. We're here in the week of October 17 to explore the latest in cloud computing and virtualization infrastructure developments.

I'm Dana Gardner, Principal Analyst at Interarbor Solutions, and I’ll be your host throughout this series of VMware-sponsored BriefingsDirect discussions. [Disclosure: VMware is a sponsor of BriefingsDirect podcasts.]

Our next case study focuses on how Germany’s largest travel agency has remade their PC landscape across 580 branch offices using virtual desktops. We’ll learn how Germany’s DER Deutsches Reisebüro redefined the desktops delivery vision and successfully implemented 2,300 Windows XP desktops as a service.

Here to tell us what this major VDI deployment did in terms of business, technical, and financial payoffs is Sascha Karbginski, Systems Engineer at DER Deutsches Reisebüro, based in Frankfurt. Welcome to the show, Sascha.

Sascha Karbginski: Hi, Dana.

Gardner: Why were virtual desktops such an important direction for you? Why did it make sense for your organization?

Karbginski: In our organization, we’re talking about 580 travel agencies all over the country, all over Germany, with 2,300 physical desktops, which were not in our control. We had life cycles out there of about 4 or 5 years. We had old PCs with no client backups.

The biggest reason is that recovery times at our workplace were 24 hours between hardware change and bringing back all the software configuration, etc. Desktop virtualization was a chance to get the desktops into our data center, to get the security, and to get the controls.

Gardner: So this seemed to be a solution that’s solved many problems for you at once.

Karbginski: Yes. That’s right.

Gardner: All right. Tell me a little bit about DER, the organization. I believe you’re a part of the REWE Group and you’re the number one travel business in Germany. Tell us a little bit about your organization before we go further into why desktop virtualization is good for you.

Karbginski: DER in Germany is the number one in travel agencies. As I said, we're talking about 580 branches. We’re operating as a leisure travel agency with our branches, Atlasreisen and DER, and also, in the business travel sector with FCm Travel Solutions.

IT-intensive business

Gardner: This is a very IT-intensive business now. Everything in travel is done though networked applications and cloud and software-as-a-service (SaaS) services. So a very intensive IT activity in each of these branches.

Karbginski: That’s right. Without the reservation systems, we can’t do any flight bookings or reservations or check hotel availability. So without IT, we can do nothing.

Gardner: And tell me about the problem you needed to solve in a bit more detail. You had four generations of PCs. You couldn’t control them. It took a lot of time to recover if there was a failure, and there was a lot of different software that you had to support.

Karbginski: Yes. We had no domain integration no control and we had those crashes, for example. All the data would be gone. We had no backups out there. And we changed the desktops about every four or five years. For example, when the reservation system needed more memory, we had to buy the memory, service providers were going out there, and everything was done during business hours.

Gardner: Okay. So this would have been a big toll on your helpdesk and for your support. With all of these people in these travel bureau locations calling you, it sounds like it was a very big problem.

There were some challenges during the rollout. The bandwidth was a big thing.



To what degree have you fully virtualized all of these desktops? Do you have a 100-percent deployment or you face deployment across these different organizations and these different agencies?

Karbginski: We have nearly about 100 percent virtualization now. We have only two or three offices, which are coming up next. We have some problem with the service provider for the VPN connection. So it's about 99 percent virtualization.

Gardner: That's pretty impressive. What were some of the issues that you encountered in order to enable this? Were there network infrastructure or bandwidth issues? What were some of the things that you had to do in order to enable this to work properly?

Karbginski: There were some challenges during the rollout. The bandwidth was a big thing. Our service provider had to work very hard for us, because we needed more bandwidth out there. The path we had our offices was 1 or 2-Mbit links to the headquarters data center. With desktop virtualization, we need a little bit more, depending on the number of the workplaces and we needed better quality of the lines.

So bandwidth was one thing. We also had the network infrastructure. We found some 10-Mbit half-duplex switches. So we had to change it. And we also had some hardware problems. We had a special multi-card board for payment to read out passports or to read out credit card information. They were very old and connected with PS/2.

A lot of problems

So there were a lot of problems, and we fixed them all. We changed the switches. Our service provider for Internet VPN connection brought us more quality. And we changed the keyboards. We don’t need this old stuff anymore.

Gardner: And so, a bit of a hurdle overcome, but what have been some of the payoffs? How has this worked out in terms of productivity, energy savings, lowering costs, and even business benefits?

Karbginski: Saving was our big thing in planning this project. The desktops have been running out there now about one year, and we know that we have up to 80 percent energy saving, just from changing the hardware out there. We’re running the Wyse P20 Zero Client instead of physical PC hardware.

Gardner: How about on the server side; are there energy benefits there?

Karbginski: We needed more energy for the server side in the data center, but if you look at it, we have 60 up to 70 percent energy savings overall. I think it’s really great.

Gardner: That’s very good. So what else comes in terms of productivity? Is there a storage or a security benefit by having that central control?

The data is under our control in the data center, and important company information is not left in an office out there.



Karbginski: As far as security, we've blocked the USB sticks now out there. So the data is under our control in the data center, and important company information is not left in an office out there. Security is a big thing.

Gardner: And how about revisiting your helpdesk and support? Because you have a more standardized desktop infrastructure now, you can do your upgrades much more easily and centrally and you can support people based on an access right directly to the server infrastructure. What’s been the story in terms of productivity and support in helpdesk?

Karbginski: In the past, the updates came during the business hours. Now, we can do all software updates at nights or at the weekends or if the office is closed. So helpdesk cost is reduced about 50 percent.

Gardner: Wow. That adds up.

Karbginski: Yeah, that’s really great.

Gardner: How big a team did it take to implement the virtualized desktop infrastructure activity for you? Was this a big expenditure in terms of people and time to get this going?

Few personnel

Karbginski: We built up the whole infrastructure -- I think it was in 9 or 10 months without the planning -- with a team of three persons, three administrators.

Gardner: Wow.

Karbginski: And now we're managing, planning, deploying, and updating it. I really think it's not a good idea to do with just three people, but it works.

Gardner: And you’ve been the first travel organization in Germany to do this, but I understand that others are following into your footsteps.

Karbginski: I've heard from some other companies that are interested in a solution like this. We were the first one in Germany, and many people told us that it wouldn't work, but we showed it works.

Gardner: And you're a finalist for the TechTarget VMware Best Award because of the way in which you’ve done this, how fast you’ve done it, and to the complete degree that you’ve done it. So I hope that you do well and win that.

We built up the whole infrastructure with a team of three persons, three administrators.



Karbginski: I received an email that we are one of the finalists, and it would be a great thing.

Gardner: Tell me now that we understand the scope and breadth of what you’ve done, a little about some of the hurdles that you’ve had to overcome. The fact that you're doing this with three people is very impressive. What does the implementation consist of? What is it you’ve got in place in terms of product that has become your de-facto industry stack for VDI?

Karbginski: I can also talk about some problems we had with this, because with the network component, for example, we have another team for it.

Gardner: I was actually wondering what products are in place? What actual technology have you chosen that then enabled you to move in this direction so well? Software, hardware, the whole stack, what is the data center stack or set of components that enables your VDI?

Karbginski: We're using Dell servers with two sockets, quad-core, 144-gigabyte RAM. We're also using EMC Clariion SAN with 25 terabytes. Network infrastructure is Cisco, based on 10 GB Nexus data center switches. At the beginning the project, we had View 4.0 and we upgraded it last month to 4.6.

The people side

Gardner: What were some of the challenges in terms of working this through the people side of the process? We've talked about process, we've talked technology, but was there a learning curve or an education process for getting other people in your IT department as well as the users to adjust to this?

Karbginski: There were some unknown challenges or some new challenges we had during the rollout. For example, the network team. The most important thing was understanding of virtualization. It's an enterprise environment now, and if someone, for example, restarts the firewall in the data center, the desktops in our offices were disconnected.

It's really important to inform the other departments and also your own help desk.

Gardner: So there are a lot of different implications across the more traditional or physical environment. How about users? Have they been more satisfied? Is there something about a virtual desktop, perhaps the speed at which it boots up, or the ability to get new updates and security issues resolved? How have the end users themselves reacted?

Karbginski: The first thing that the end users told us was that the selling platform from Amadeus, the reservation system, runs much faster now. This was the first thing most of the end users told us, and that’s a good thing.

The next is that the desktop follows the user. If the user works in one office now and next week in another office, he gets the same desktop. If the user is at the headquarters, he can use the same desktop, same outlook, and same configuration. So desktop follows the user now. This works really great.

The desktop follows the user. If the user works in one office now and next week in another office, he gets the same desktop.



Gardner: Looking to the future, are you going to be doing this following-the-user capability to more devices, perhaps mobile devices or at home PCs? Is there the ability to take advantage of other endpoints, perhaps those even owned by the end users themselves and still deliver securely the applications and data that you need?

Karbginski: We plan to implement the security gateway with PCoIP support for home office users or mobile users who can access their same company desktop with all their data on it from nearly every computer in the world to bring the user more flexibility.

Gardner: So I should think that would be yet another payoff on the investments that you’ve made is that you will now be able to take the full experience out to more people and more places, but for relatively very little money to do that.

Karbginski: The number of desktops is still the same, because the user gets the same desktop. We don’t need for one user two or three desktops.

Gardner: Right, but they're able to get the information on more devices, more screens as they say, but without you having to buy and manage each of those screens. How about advice for others? If you were advising someone on what to learn from your experience as they now move toward desktop virtualization, any thoughts about what you would recommend for them?

Inform other departments

Karbginski: The most important thing is to get in touch with the other departments and inform them about the thing you're doing. Also, inform the user help desk directly at the beginning of the project. So take time to inform them what desktop virtualization means and which processes will change, because we know most of our colleagues had a wrong understanding of virtualization.

Gardner: How was it wrong? What was their misunderstanding do you think?

Karbginski: They think that with virtualization, everything will change and we'll need other support servers, and it's just a new thing and nobody needs it. If you inform them what you're doing that nothing will be changed for them, because all support processes are the same as before, they will accept it and understand the benefits for the company and for the user.

The most important thing is to get in touch with the other departments and inform them about the thing you're doing.



Gardner: We’ve been talking about how DER Deutsches Reisebüro has been remaking their PC landscape across 500 in 80 branch officers. They're part of Germany’s largest travel agency and they’ve been deploying desktop virtualization successfully, but very broadly up to 100 percent across their environment. So it's very impressive. I’d like to thank our guest, Sascha Karbginski, Systems Engineer there at DER Deutsches Reisebüro. Thank you so much, Sascha.

Karbginski: Thank you, Dana.

Gardner: And thanks to our audience for joining this special podcast coming to you from the VMworld 2011 Conference in Copenhagen. I'm Dana Gardner, Principal Analyst at Interarbor Solutions, your host throughout this series of VMware-sponsored BriefingsDirect discussion. Thanks again for listening, and come back next time.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: VMware.

Transcript of a sponsored podcast discussion from VMworld 2011 in Copenhagen on how DER Deutsches Reisebüro virtualized 2,300 desktops to centralize administration. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in:

Tuesday, November 08, 2011

Case Study: Southwest Airlines' Productivity Takes Off Using Virtualization and IT as a Service

Transcript of a BriefingsDirect podcast on how travel giant Southwest Airlines is using virtualization to streamline customer service applications.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: VMware.

Dana Gardner: Hello, and welcome to a special BriefingsDirect podcast series coming to you in conjunction with a recent VMworld 2011 Conference.

I'm Dana Gardner, Principal Analyst at Interarbor Solutions, and I’ll be your host throughout this series of VMware-sponsored BriefingsDirect discussions.

Our next VMware case study interview focuses on Southwest Airlines, one of the best-run companies anywhere, with some 35 straight years of profitability, and how "IT as a service" has been transformative for them in terms of productivity.

Here to tell us more about how Southwest is innovating and adapting with IT as a compelling strategic differentiator is Bob Young, Vice President of Technology and Chief Technology Officer at Southwest Airlines. [Disclosure: VMware is a sponsor of BriefingsDirect podcasts.]

Welcome to BriefingsDirect, Bob.

Bob Young: Well, thank you very much. I appreciate the opportunity to speak with you.

Gardner: We have heard a lot about IT as a service, and unfortunately, a lot of companies face an IT organization that might be perceived as a little less than service-oriented, maybe even for some a roadblock or a hurdle. How have you at Southwest been able to keep IT squarely in the role of enablement?

Young: First off, as everybody should know already, Southwest Airlines is the customer service champ in the industry. Taking excellent care of our customers is just as important as filling our planes with fuel. It’s really what makes us go.

So as we are taking a look and trying to be what travelers want in an airline, and we are constantly looking for ways to improve Southwest Airlines and make it better for our customers, that's really where virtualization and IT as a service comes into play. What we want to be able to do is make IT not say, "Oh, this is IT versus something else."

People want to be able to get on Southwest.com, make a reservation, log on to their Rapid Rewards or our Loyalty Program, and they want to be able to do it when they want to do it, when they need to do it, from wherever they are. And it’s just great to be able to provide that service.

We provide that to them at any point in time that they want in a reliable manner. And that's really what it gets right down to -- to make the functions and the solutions that we provide ubiquitous so people don’t really need to think about anything other than, "I need to do this and I can do it now."

At your fingertips

Gardner: I travel quite a bit and it seems to me that things have changed a lot in the last few years. One of the nice things is that information seems to be at your fingertips more than ever. I never seem to be out of the loop now as a traveler. I can find out changes probably as quickly as the folks at the gate.

So how has this transfer of information been possible? How have you been able to keep up with the demands and the expectations of the travelers?

Young: One of the things that we like to do at Southwest Airlines is listen to our customers, listen to what their wants and desires are, and be flexible enough to be able to provide those solutions.

If we talk about information and the flow of information through applications and services, it really starts to segment the core technical aspects of that so the customer and our employees don’t really need to think about it. When they want to get the flight at the gates, the passenger is on a flight leg, etc., they can go ahead and get that at any moment in time.

Another good example of that is earlier this year we rolled out our new Rapid Rewards 2.0 program. It represents a bold and leading way to look at rewards and giving customers what they want. With this program, we've been able to make it such that we can make any seat available on any flight for our Rapid Rewards customers for rewards booking, which is unique in the industry.

What we want to be able to do is provide it whenever they want it, whenever they need it, at the right cost point, and to meet their needs.



The other thing it does is allows our current and potential members the flexibility in how they both earn miles and points and how they use them for rewards -- being able to plan ahead and allowing them to save some significant points.

The same is true of how we provide IT as a service. What we want to be able to do is provide it whenever they want it, whenever they need it, at the right cost point, and to meet their needs. We've got some of the best customers in the world and they like to do things for themselves. We want to allow them to do that for themselves and be able to provide our employees the same areas.

If you've been on a Southwest flight, you've seen our flight crews, our in-flight team, really trying to have fun and trying to make Southwest a fun place to work and to be, and we just want to continue to support that in a number of different ways.

Gardner: You have also had some very significant challenges. You're growing rapidly. Your Southwest.com website is a crucial and growing part of your revenue stream. You've had mergers, acquisitions, and integrations as a result of that, and as we just discussed, the expectations of your consumers, your customers, are ramping up as well -- more data, more mobile devices, more ways to intercept the business processes that support your overall products and services.

So with all those requirements, tell me a little bit about the how. How in IT have you been able to create common infrastructures, reduce redundancy, and then yet still ramp up to meet these challenging requirements?

Significant volume

Young: As you all know, Southwest.com is a very large travel site, one of the largest in the industry -- not just airlines, but the travel industry as a whole. Over 80 percent of our customers and consumers book travel directly on Southwest.com. As you may know, we have fare sales a couple of times a year, and that can drive a significant volume.

What we've been able to do and how we have been able to meet some of those challenges is through a number of different VMware products. One of the core products is VMware itself, if we talk about vSphere, vMotion, etc., to be able to provide that virtualization. You can get a 1-to-10 virtualization depending on which type of servers and blades you're using, which helps us on the infrastructure side of the house to maintain that and have the storage, physical, and electrical capacity in our data centers.

But it also allows us, as we're moving, consolidating, and expanding these different data centers, to be able to move that virtual machine (VM) seamlessly between points. Then, it doesn’t matter where it’s running.

That allows us the capacity. So if we have a fare sale and I need to add capacity on some of our services, it gives our us and our team that run the infrastructure the ability to bring up new services on new VMs seamlessly. It plugs right into how we're doing things, so that internal cloud allows us not to experience blips.

It's been a great add for us from a capacity management perspective and being able to get the right capacity, with the right applications, at the right time. It allows us to manage that in such a way that it’s transparent to our end-users so they don’t notice any of this is going on in the background, and the experience is not different.

It's been a great add for us from a capacity management perspective and being able to get the right capacity, with the right applications, at the right time.



Gardner: I understand that you're at a fairly high level of virtualization. Is that a place where you plan to stay? Are you going to pursue higher levels? Where do you expect to go with that?

Young: I'll give you a little bit of background. We started our virtualized environments about 18 months ago. We went from a very small amount of virtualization to what we coined our Server 2.0 strategy, which was really the combination of commodity-based hardware blades with VMware on that.

And that allowed us last year in the first and second quarter to grow from several hundred VMs to over several thousand, which is where we're at today in the production environment. If you talk about production, development, and test, production is just one of those environments.

It has allowed us to scale that very rapidly without having to add a thousand physical servers. And it has been a tremendous benefit for us in managing our power, space, and cooling in the data center, along with allowing our engineers who are doing the day-to-day work to have a single way to manage it, deploy, and move stuff around even more automatically. They don’t have to mess with that anymore, VMware just takes care of the different products that are part of the VMware Suite.

Gardner: And your confidence, has it risen to the level where you're looking at 70, 80, 90, even more percent of virtualization? How do you expect to end that journey?

Ready for the evolution

Young: I would love to be at 100 percent virtualized. That would be fantastic. I think unfortunately we still have some manufacturers and software vendors -- and we call them vendors, because typically we don’t say partners -- who decide they are not going to support their software running in the virtualized environment. That can create problems, especially when you need to keep some of those systems up 24 x 7, 365, with 99.95 percent availability.

We're hoping that changes, but the goal would be to move as much as we can, because if I take a look at virtualization, we are kind of our internal private cloud. What that’s really doing is getting us ready for the evolution that’s going to happen over the next, 5, 7, or 10 years, where you may have applications and data deployed out in a cloud, a virtual private cloud, public cloud if the security becomes good enough, where you've got to bring all that stuff together.

If you need to have huge amounts of capacity and two applications are not co-located that need to talk back and forth, you've got to be much more efficient on the calls and the communications and make that seamless for the customer.

This is giving us the platform to start learning more and start developing those solutions that don’t need to be collocated in a data center or in one or two data centers, but can really be pushed wherever it makes sense. That could be from wherever the most efficient data center is from a green technology perspective, use the least electricity and cooling power, to alternate energy, to what makes sense at the time of the year.

That is a huge add and a huge win for us in the IT community to be able to start utilizing some of that virtualization and even across physical locations.

It allows us to deploy that stuff within minutes, whereas it used to take engineers manually going to configure each thing separately. That’s been a huge savings.



Gardner: So as you've ramped up on your virtualization, I imagine you have been able to enjoy some benefits from that in terms of capital expense, hardware, and energy. How about in some of the areas around configuration management and policy management. Is there a centralization feature to this that also is paying dividends?

Young: That’s a huge cornerstone of the suite of tools that we've been able to get through VMware is being able to deploy custom solutions and even some of the off-the-shelf solutions on a standard platform, standard operating systems, standard configurations, standard containers for the web, etc. It allows us to deploy that stuff within minutes, whereas it used to take engineers manually going to configure each thing separately. That’s been a huge savings.

The other thing is, once you get the configuration right and you have it automated, you don’t have to worry about people taking some human missteps. Those are going to happen, and you've got to go back and redo something. That elimination of error and the speed at which we can do that is helping. As you expand your server footprints and the number of VMs and servers you have without having to add to your staff, you can actually do more with the same number of or fewer staff.

Gardner: I wonder how you feel about desktop virtualization. Another feature that we've seen in the field in the marketplace is those that make good use of server virtualization are in a better position to then take that a step further and extend it out through PC-over-IP and other approaches to delivering the whole desktop experience. Is that something that you're involved with or experimenting with? How do you feel about that?

Young: This has been going on and off in the IT industry for the past 10-15 years, if you talk about Net PCs and some of the other things. What’s really driven us to take a look at it is that around our environment we can control security on virtual desktops, etc., very clearly, very quickly and deliver that in a great service.

New mobile devices

The other thing that’s leading to this is, not just what we talked about in security, is the plethora of brand new mobile devices -- iPhones, iPads, Android devices, Galaxy. HP has a new device. RIM has a new device. We need to be able to deliver our services in a more ubiquitous manner. The virtual desktop allows us to go ahead and deliver some of those where I don’t need to control the hardware. I just control the interface, which can protect our systems virtually, and it’s really pretty neat.

I was on one of my devices the other day and was able to go in via virtual desktop that was set up to be able to use some of the core systems without having all that stuff loaded on my machine, and that was via the Internet. So it worked out phenomenally well.

Now, there are some issues that you have to do depending on whether you're doing collocation and facility, but you can easily get through some of that with the right virtualization setup and networking.

Gardner: So you have come an awfully long way. You say 18 months ago you were only embarking on virtualization, but now you're already talking about hybrid clouds and mobile enablement and wide area network optimization. How is that you have been able to bite off so much so soon? A lot of people would be intimidated, do more of that crawl-walk-run, with the emphasis on the crawl and walk parts?

Young: Well, I am very fortunate. I might come up with the vision of where we want to go and this is where IT is going, and I am very fortunate to have some very good and phenomenal engineers working on this, working through it, all the issues, all the little challenges that pop up along the way in order to do this.

It’s what our team would say is pretty cool technology, and it gets them excited about doing something new and different as well. I've got a couple of managers -- Tim Pilson, Mitch Mitchell -- and their teams, and some really good people.

So I really have to give credit to the teams that are working with me, my team who gets it done, and VMware for providing such a great product.



Jason Norman is one of the people, and Doug Rowland also has been very involved with getting this rolled out. It’s amazing what a core set of just a few people can do with the right technology, the right attitude, and passion to get it done. I've just been very impressed with their, what we call warrior spirit here at Southwest Airlines -- just not giving up, doing what it takes to get it done, and being able to utilize that with some of the VMware products.

It extends beyond that team. Our development teams use Spring and some other of the VMware products as well. If we run into an issue, it’s just like VMware on the development side of the house and product side of the house is really part of our extended team. They take it, they listen, and they come back with a fix and a patch in literally a day or two, rather than some other vendors with whom you might wait weeks or months and it might never make it to you.

So I really have to give credit to the teams that are working with me, my team who gets it done, and VMware for providing such a great product that the engineers want to use it, can use it, and can understand it, and make huge amounts of progress in a very short period of time.

Gardner: Well, great. It’s a very interesting and compelling story. We've been talking with Southwest Airlines and how they are continuing to innovate and adapt and using IT as a compelling strategic differentiator.

Our guest has been Bob Young, Vice President of Technology and Chief Technology Officer at Southwest Airlines. Thanks so much, Bob.

Young: Well, thank you.

Gardner: I'm Dana Gardner, Principal Analyst at Interarbor Solutions, your host throughout this series of VMware-sponsored BriefingsDirect discussions. Thanks again for listening, and come back next time.

Listen to the podcast. Find it on iTunes/iPod. Download the transcript. Sponsor: VMware.

Transcript of a BriefingsDirect podcast on how travel giant Southwest Airlines is using virtualization to streamline customer service applications. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in: