Tuesday, July 19, 2011

Cloud and SaaS Force a Rethinking of Integration and Middleware as Services for Services

Transcript of a BriefingsDirect podcast of the role of cloud and SaaS in the changing landscape of application integration.

Listen to the podcast. Find it on iTunes/iPod and Podcast.com. Download the transcript. Sponsor: Workday.

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 major trends around cloud, mobile, and software as a service (SaaS) are dramatically changing the requirements and benefits of application integration.

In many respects, the emphasis now on building hybrid business processes from a variety of far-flung sources forces a rethinking of integration and middleware. Integration capabilities themselves often need to be services in order to support a growing universe of internal and external constituent business process component services.

Here to explore the new era of integration-as-a-service and what it means for the future is David Clarke, Director of Integration at SaaS ERP provider Workday, and who is based in Dublin. Welcome, David. [Disclosure: Workday is a sponsor of BriefingsDirect podcasts.]

David Clarke: Hi, Dana. Good to be here.

Gardner: As I said, the past is necessarily prologue when it comes to integration. Why have the platforms, applications, and data of the past forced a certain approach to integration, and why is that ill-suited to what we are expecting and seeing more of every day with cloud and SaaS?

Clarke: One thing is that historically applications were built, structured, and architected in very different ways. When you tried to knit those things together, there was quite a diverse set of requirements, which needed to be addressed. And, there was a very wide variation in their architectures. So that implied a very general-purpose middleware that had to cope with many different and very diverse scenarios. That was one factor.

A second factor was that the middleware and the integration tended to come as an afterthought. So, it couldn't really influence or inform the way that the application platforms themselves were designed.

Those two things together made it more difficult than it needed to be. Then, what we're starting to see -- and what we are certainly hopeful of seeing in this generation built around cloud -- is that those two things aren’t necessarily the case. So, we can benefit more from having integration designed in upfront, and having a more consistent overall architecture, so that it’s essentially easier for it to plug into.

One third variable might be that customers historically also discounted or underestimated the likely impact or complexity of doing integration. They tended to come to them as an afterthought and then struggled with them. This time around, to some extent, they've been burned by previous generations. So they're more wary, and are additionally including integration more in their upfront planning.

Gardner: We're not necessarily talking about throwing the baby out with the bath water here. We're still going to be doing integrations in the traditional way. It's just that we need to add another category, and it seems that there is a benefit in that. We can use many of the tools, many of the underlying technologies that supported traditional middleware, and extend that into this services environment.

Clarke: Correct. There’s nothing fundamentally new, in some sense. I've worked in several generations of integration and middleware technology, and each one is a refinement of the past, and you're standing on the shoulders of giants, to badly paraphrase Newton.

Packaged and presented

A
lot of the underlying technology you're using for integration, a lot of the underlying concepts, are not that new. It's just the way that they're being packaged and presented. In some cases, it's the protocols that we're using, and certainly some of the use models. But the ways you're accessing them and consuming them are different. So, it is in that sense [this is] evolutionary.

Gardner: What has probably changed the most are the requirements. The problem set that we're addressing has changed. How has it changed? Perhaps this would be an opportunity for you as well to explain what Workday is, what it does, and how you came to be a part of the Workday team.

Clarke: Historically, integration technology was sold as a stand-alone and on-premise offering. Companies would buy or build applications, and then, as their business processes evolved, they found a need to integrate them and connect them together. So, they would license middleware and use that to achieve that.

There have been a couple of generations of middleware technology companies that have helped customers do this. They shared some of the characteristics around certain generations of technology.

So you had companies like TIBCO Software in the early days, folks from the financial sectors. Then you had companies like BEA Systems focused on the Java generation of middleware and application servers. And then, you had more XML and web services-centric companies.

As a middleware vendor, you're trying to solve essentially any and every problem.



You had those three generations, but what was common to them all was that they were quite divorced from the application experience. And that was my background in pure middleware, building and selling that technology.

A strength and a weakness of that was that it was very general purpose. As a middleware vendor, you're trying to solve essentially any and every problem. To draw on the Eclipse Foundation's motto: middleware is a general-purpose platform that can do everything or nothing. In many cases, people ended up spending a lot of money on this general-purpose middleware and essentially achieving nothing, which was frustrating.

Workday is an applications company. We're an on-demand apps company and we build and serve human capital management (HCM), financials, and enterprise resource planning (ERP) application suites.

Cape Clear, which was my former company, was acquired by Workday about three years ago. We were partners, but as Workday’s business expanded significantly, they saw that providing a compelling and a differentiated integration experience in the context of this new cloud architecture was going to be something that was very important to them. So they acquired Cape Clear and we became part of the overall Workday organization.

The first surprise to me was that I had always worked for companies where it was difficult essentially to explain what we did. You couldn't really go to your grandmother and describe middleware technology, whereas you could at least go and explain financial systems or HCM systems.

Overarching context

That then flowed all the way down through to how we positioned, thought about, described, and marketed the technology. It has certainly been my experience that it's a lot easier to describe, position, plan, and explain integration technology when you have this overarching context of an application domain.

That's been very instructive and has interesting implications in the future for the nature, or indeed the existence, of a stand-alone in the middleware market. That might be an interesting topic we could talk about later.

The other observation is that the consistency of these use cases make our jobs somewhat easier. What's also been interesting is the nature of the load and the scale profiles that we're seeing.

A lot of middleware applications that we used to see were technically complex to achieve, but were often relatively low volume or relatively low scale. But, in large-scale companies, when you're dealing with their core systems of record around financials and HCM, you're looking at very large data sets, with very significant scalability requirements, and very significant performance constraints. That has interesting implications for how you think about and implement your middleware solutions.

Gardner: So there seem to be two fundamental things going on here. One, is taking integration to the on-demand or SaaS domain, but second, there is also this embedding integration functionality into the application.

One of the perpetual holy grails of the middleware industry, when it was a stand-alone undertaking was to find a way to express and expose middleware and integration concepts in a way that they could be used by mere mortals.



People, when they use Workday -- whether they are human resources professionals or employees in these organizations, whether they're partners or suppliers to these enterprises that are using Workday -- they're not thinking about integration. They're thinking about human resources, benefits, payroll, and insurance.

Tell me how this shift to on-demand, as well as embedding into the application, changes the requirements. How does someone like yourself who is crafting the middleware integration capabilities need to shift their thinking in order to go “to the cloud,” but also become part-and-parcel with the application?

Clarke: One of the perpetual holy grails of the middleware industry, when it was a stand-alone undertaking, was to find a way to express and expose middleware and integration concepts in a way that they could be used by mere mortals, by business analysts, by people who weren't necessarily very deep technologists with deep technology expertise.

In my experience, the middleware industry never achieved that. So, they didn't really ever find a metaphor or a use model that enabled less skilled, but nonetheless technically savvy, people to use their products.

As you observe in the applications game, you absolutely have to get there, because fundamentally what you're doing here is you are enabling companies and individuals to solve business problems and application problems. The integration arises as a necessity of that or as a consequence of that. In and of itself, it isn't useful.

Designing applications

The most specific thing that we've seen is how we build, manipulate, and use extremely sophisticated integration technology. We spend a lot of our time thinking about how to design that into the application, so that it can be experienced and consumed by users of the application who don’t know anything about XML, Java, web protocols, or anything like that.

To give you one very simple example, the most common use case of all probably is people getting data, perhaps from our system, doing something with it -- a simple transformation -- and then delivering it or putting it somewhere else, perhaps into our system.

That model of "get, transform, and put" is intuitively straightforward, but historically that has always been realized in a complicated way in the middleware stack. We've built a very simple tool inside of our application, and it's now the most heavily used integration component in our system.

Business analysts can very easily and visually define what they are getting and putting it in terms of the business concepts and the business objects they understand. They can define very simple transformations, for example, going from a payroll input to a check, or going from a report of absences by departments to a payroll input.

They're consuming and using integration technologies in a very natural way in the context of their day-to-day working in the web layer in these systems. They're not programmers. They're not developers. They're not thinking about it that way.

It's quite empowering for the teams that we have had working on this technology to see if it's usable in that way by the business analysts here. It's the closest I've seen people get to capturing this unicorn of enabling integration technology to be actually used by business people.

It's the closest I've seen people get to capturing this unicorn of enabling integration technology to be actually used by business people.



Gardner: While you have put quite a bit of emphasis on the tool side in order to make this something that mere mortals can adjust and operate, you've also done a lot of heavy lifting on the connections side. You recognized that in order to be successful with an integration platform, you had to find the means in which to integrate to a vast variety of different types of technologies, services, data, and so forth. Tell me what you've done, not only on the usability, but on the applicability across a growing universe of connection points.

Clarke: That’s another interesting area. As you say, there are thousands or millions of different types of endpoints out there. This being software, it can map any data format to any other data format, but that’s a trivial and uninformative statement, because it doesn’t help you get a specific job done.

Essentially what we've been trying to do is identify categories of target systems and target processes that we need to integrate with and try to optimize and focus our efforts on that.

For example, pretty much the majority of our customers have a need to integrate to and from benefit systems for 401(k), healthcare, dental, visual plans, and so forth. It's an extremely common use case. But, there is still a wide diversity of benefits providers and a wide variety of formats that they use.

We've studied the multiple hundreds of those benefits providers that we've experienced by working with our customers and we've abstracted out the most common format scenarios, data structures, and so forth, and we have built that into our integration layer.

Configure your data set

You can very easily and rapidly and without programming configure your specific data set, so that it can be mapped into and out of your specific set of benefits providers, without needing to write any code or build a custom integration.

We've done that domain analysis in a variety of areas, including but not limited to benefits. We've done it for payroll and for certain kinds of financial categories as well. That's what's enabling us to do this in a scalable and repeatable way, because we don’t want to just give people a raw set of tools and say, "Here, use these to map anything to anything else." It's just not a good experience for the users.

Gardner: David, you mentioned that Cape Clear was acquired by Workday about three years ago, and Workday has been growing very rapidly. Have you been surprised by the adoption rate and pattern around SaaS, and now we're talking about cloud and hybrid cloud? Did this happen faster than you were expecting, because it certainly caught me by surprise.

Clarke: Totally. I remember when we originally became part of Workday several years ago, we were doing some sort of product planning and strategic thinking about how we were going to integrate the product lines and position them going forward. One of the things we had in our roadmap at the time was this idea of an appliance. So we said, "Look, we can envision the future, where all the integration is done in the cloud, but we frankly think it's like a long way off. We think that it's some years off."

For that reason, we articulated and embarked on a path of offering what we were calling an appliance, which essentially would have been an on-premise component to the integration stack or of the integration stack that would be deployed at customer sites. We thought the world wasn’t going to be ready soon enough to put the integration technology and stack in the cloud as well.

It just became clearer and clearer to us that there was an appetite and a willingness in our customer and prospect base to use this technology in the cloud.



Happily that turned out to have been incorrect. Over the course of the ensuing 12 months, it just became clearer and clearer to us that there was an appetite and a willingness in our customer and prospect base to use this technology in the cloud.

We never really went ahead with that appliance concept, it didn’t get productized. We never used it. We don’t need to use it. And now, as I have conversations with customers and with prospects, it just is not an issue.

In terms of it being any kind of philosophical or in principle difficulty or challenge, it has just gone away. It totally surprised me, as well, because I expected it to happen, but thought it would take a lot longer to get to where it has got to already.

Gardner: There is a certain irony, because we were all involved with service-oriented architecture (SOA) and kept waiting for that to get traction, and were a little bit distressed that it wasn’t catching on. Then, lo and behold, this concept of SaaS and cloud leapfrogs and catches on much faster than we thought. So, it is an interesting time.

When we go back to enterprises, we recognize that this “consumerization” of IT is taking place, where the end-users, the zeitgeist of expectations, is now at the point where they want IT in the enterprise to work as well and in the same manner as it does for their personal lives. How does that shift the thinking of an enterprise architect?

Clarke: Superficially, enterprise architects are under a lot of pressure to, as you say, to present technologies in ways that are more familiar to customers from their personal lives. The most specific example of that is the embrace of mobile technologies. This isn't a huge surprise. It's been a pretty consistent pattern over a number of years that workforce mobility is a major influence on product requirements.

Mobile devices

We've seen that very significant proportions of access to our system is via mobile devices. That informs our planning and our system architecture. We're invested heavily in mobile technologies -- iPad, Android, BlackBerry, and other clients. In my experience, that’s something that's new, with the customer enterprise architects. This is something they have to articulate, defend, and embrace.

Historically, they would have been more concerned with the core issues of scalability, reliability, and availability. Now, they've got more time to think about these things, because we as SaaS vendors have taken a lot of things that they used to do off of their plates.

Historically, a lot of time was spent by enterprise architects worrying about the scalability and reliability of the enterprise application deployments that they had, and now that’s gone away. They get a much higher service level agreement (SLA) than they ever managed to operate by themselves when they run their own systems.

So, while they have different and new things to think about because of the cloud and mobility, they also have more head space or latitude to do that, because we have taken some of the pain that they used to have away.

Gardner: I suppose that as implications pan out around these issues, there will be a shift in economics as well, whereby you would pay separately and perhaps on a capital and then operating basis for integration.

They also have more headspace or latitude to do that, because we have taken some of the pain that they used to have away from them.



If integration by companies like Workday becomes part-and-parcel of the application services -- and you pay for it on an operating basis only -- how do traditional business models and economics around middleware and integration survive? How do you see this transition working, not only for the functionality and the architecture, but in dollars and cents?

Clarke: I'd certainly hate to be out there trying to sell middleware offerings stand-alone right now, and clearly there have been visible consolidations in this space. I mentioned BEA earlier as being the standard bearer of the enterprise Java generation of middleware that’s been acquired by Oracle.

They are essentially part of the application stack, and I'm sure they still sell and license stand-alone middleware. Obviously, the Oracle solutions are all on-premise, so they're still doing on-premise stuff at that level. But, I would imagine that the economics of the BEA offering is folded very much into the economics of the Oracle application offering.

In the web services generation of middleware and integration, which essentially came after the enterprise Java tier, and then before the SOA tier, there was a pretty rapid commoditization. So, this phenomenon was already starting to happen, even before the cloud economics were fully in play.

Then, there was essentially an increased dependence or relevance of open source technologies -- Spring, JackBe, free stacks -- that enabled integration to happen. That commoditization was already starting to happen.

Open source pressure

So even before the advent of the cloud and the clear economic pressure that put on stand-alone integration, there was already a separate pressure that was originating from open source. Those two things together have, in my view, made it pretty difficult to sustain or to conceive a sustainable integration model.

A lot of the investment dollars that have gone into something like integration market are now going elsewhere in infrastructure. They're going into storage. They're going into availability. They're going certainly to cloud platforms. It would need to be a brave venture capitalist now who would write a check to a company coming in with a bright idea for a new on-premise middleware stack. So that business is gone.

Gardner: We're also seeing some investment around taking open source middleware and integration capabilities and extending them to the cloud. It's not as difficult for an open source company, because their monetization has been around maintenance and support, more of an operating expense. We certainly haven’t seen too much in the way of a general-purpose integration cloud from any of the traditional on-premises middleware vendors.

Do you think in 10 years, or maybe 5, we won’t even be thinking about integration? It will really be a service, a cloud service, and perhaps it will evolve to be a community approach. Those people who need to be connected to one another will either structurally move toward some standardization or, perhaps in a more ad hoc or organic way, provide the means by which they could more easily play well together?

Clarke: There are a couple of things that we see happening here. I'll make two main observations in this area.

There is an important difference between a general-purpose platform or integration platform and then a more specific one, which is centered around a particular application domain. Workday is about the latter.



First, at the risk of losing half our audience with the jargon, there is an important difference between a general-purpose platform or integration platform and then a more specific one, which is centered around a particular application domain. Workday is about the latter.

We're building a very powerful set of cloud technologies, including an integration cloud or an integration platform in the cloud, but it’s very focused on connecting essentially to and from Workday, and making that very easy from a variety of places and to a variety of places.

What we're not trying to create is a general-purpose platform, an associated marketplace, in the way that maybe somebody like Salesforce.com is doing with AppExchange or Google with App Engine for app development. In a sense, our scope is narrower in that way, and that’s just how we're choosing to prosecute the opportunity, because it’s harder to establish a very horizontal platform and it’s just difficult to do.

I referred earlier to the problem that middleware companies traditionally have of doing everything and nothing. When you have a purely horizontal platform that can offer any integration or any application, it’s difficult to see exactly which ones are going to be the ones that get it going.

The way we're doing this is therefore more specific. We have a similar set of technologies and so on, but we're really basing it very much around the use case that we see for Workday. It’s very grounded in benefits integrations, payroll integrations, financial integrations, payment integrations. And every one of our deployments has tens, dozens, hundreds of these integrations. We're constantly building very significant volume, very significant usage, and very significant experience.

Developing marketplace

I can see that developing into a marketplace in a limited way around some of those key areas and possibly broadening from there.

That's one of the interesting areas of distinction between the strategies of the platform vendors as to how expansive their vision is. Obviously expansive visions are interesting and creating horizontal platforms is interesting, but it’s more speculative, it’s riskier, and it takes a long time. We are more on the specific side of that.

You mentioned collaborating and how this area of business processes and people collaborating in the community. I referred earlier to this idea that we're focusing on these key use cases. What’s arising from those key use cases is a relatively small set of documents and document formats that are common to these problem areas.

Lately, I've been reading, or rereading, some of the RosettaNet stuff. RosettaNet has been around forever. It was originally created in the early '80s. As you know, it was essentially a set of documents, standard documents, interchange formats for the semiconductor or the technology manufacturing industry, and it has been very successful, not very prominent or popular, but very successful.

What we see is something similar to RosettaNet starting to happen in the application domain where, when you are dealing with payroll providers, there is a certain core set of data that gets sent around. We have integrated to many dozens of them and we have abstracted that into a core documentary that reflects the set of information and how it needs to be formatted and how it needs to be processed.

These are very good vectors for cooperation and for collaboration around integrations, and they're a good locus around which communities can develop standardized documents.



In fact, we now have a couple of payroll partners who are directly consuming that payroll format from us. So, in the same way that there are certain HR XML standards for benefits data, we can see other ones emerging in other areas of the application space.

These are very good vectors for cooperation and for collaboration around integrations, and they're a good locus around which communities can develop standardized documents, which is the basis for integration. That’s intriguing to me, because it all derives from that very specific set of use cases that I just never really saw as a general-purpose integration vendor.

Gardner: Getting back to adoption patterns and economics, it seems as if what you are proposing, and what Workday is supporting, is this application-level benefit. A business process, like a network, is perhaps more valuable as the number of participants in the process increases, and become able to participate with a fairly low level of complexity and friction.

It's sort of a derivative of Metcalfe's Law, but at the business process level, which is quite different than trying to corral an integration community around a specific platform with the intent of getting more people on that platform and having a long-term flow of license revenue as a result.

So, if we make this shift to a Metcalfe's law-type of "the more participants, the more valuable it is to all of those participants," shouldn’t we expect a little bit of a different world around integration in the next few years?

Business process

Clarke: That’s right, because of the distinction you mentioned. We don’t really see or envisage this very transactional marketplace, where you just have people buying a round of maps or integrations and installing them. We see it happening in the context of a business process.

For example, hiring. As somebody hires somebody into Workday, there are typically many integration points in that business process -- background checking, provisioning of security cards, and creation of email accounts. There is a whole set of integration points. We're increasingly looking to enable third parties to easily plug-in into those integration points in a small way, for provisioning an email account, or in a big way, like managing a whole payroll process.

It’s that idea of these integrations as being just touch points and jumping-off points from an overall business process, which is quite a different vision from writing cool, stand-alone apps that you can then find and store from inside of our platform marketplace.

It’s that idea of an extended business process where the partners and partner ISVs and customers can collaborate very easily, and not just at install time or provisioning time, but also when these processes are running and things go wrong, if things fail or errors arise.

You also need a very integrated exception handling process, so that customers can rapidly diagnose and correct these errors when they arise. Then, they have a feeling of being in a consistent environment and not like a feeling of having 20 or 30 totally unrelated applications executing that don’t collaborate and don’t know about each other and aren’t executing the context within the same business process. We're keen to make that experience seamless.

You also need a very integrated exception handling process, so that customers can rapidly diagnose and correct these errors when they arise.



Gardner: I can also see where there is a high incentive for the participants in a supply chain or a value chain of some sort to make integration work. So perhaps there is an incentive toward cooperation in ways that we hadn’t seen before. I am thinking of, at least in the human resources field, where it’s in my best interest as an insurance company or as a payroll benefits provider, for example, to work with the SaaS or cloud provider in this regard -- to the betterment of our mutual end users.

Do you already see that the perception of cooperation for integration is at a different plane? Where do you expect that to go?

Clarke: Totally, already. Increasingly -- pick an area, but let's say for learning management or something -- if we integrate, or if multiple people integrate to us or from us, then customers already are starting to expect that those integrations exist.

Now they're starting to ask about how good they are, what's the nature of them, what SLAs can they expect here? The customers are presuming that an integration, certainly between Workday and some other cloud-based service, either exists already or is very easy and doable.

But they're looking through that, because they're taking the integration technology level questions for granted. They're saying, "Given that I can make such an integration work, how is it really going to work, what's the SLA, what happens if things go wrong, what happens when things fail?"

What's really interesting to me is that customers are increasingly sophisticated about exploring the edge cases, which they have seen happen before and have heard about them before. They're coming to us upfront and saying, "What happens if I have issues when my payroll runs? Who do I go to? How do you manage that? How do you guys work with each other?"

Consistent information

We, therefore, are learning from our customers and we're going to our ISV and services partners, like our payroll partners, our learning management partners, our background checking partners and saying, "Here is the contract that our customers expect. Here is the service that they expect." They're going to ask us and we want to be able to say that this partner tests against every single update and every single revision of the Workday software. They will handle a seamless support process where you call one number and you get a consistent set of information.

Customers are really looking through the mere fact of a technical integration existing and asking about what is my experience going to be and actually using this day-to-day across 50 geographies and across population of 20,000 employees. I want that to be easy.

It’s a testament to the increasing sophistication of the integration technology that people can take that for granted. But as I say, it’s having these increasingly interesting and downstream effects in terms of what people are expecting from the business experience of using these integration systems in the context of a composite business process that extend beyond just one company.

Gardner: Moving toward closing up our conversation, David, you have raised the issue here about that one throat to choke, if you will. When you have a massive, complex, integration landscape, does it makes sense to focus on the application provider as that point of responsibility and authority? Or does it have to be federated?

Have you seen any models emerge, something that we probably could not have predicted but needs to happen on its own, in a real world setting that indicates how that issue of trust and authority might pan out?

Clarke: What we are gradually feeling our way toward here is that for us that’s the central concept of this federation of companies. We think obviously of Workday being in the middle of that. It depends on what your perspective is, but you have this federation of companies collaborating to provide the service ultimately, and the question is, where do they choke?

And it's not realistic to say that you can always come to Workday, because if we are integrating to a payroll system on behalf of somebody else, and we correctly start off and run the payroll or send the payroll requests, and then there is an error at the other end, the error is happening ultimately in the other payroll engine. We can't debug that. We can't look at what happened. We don't necessarily even know what the data is.

As we run any integration in our cloud, there is a very consistent set of diagnostics, reporting, metrics, error handling, error tracking that is generated and that's consistent.



We need a consistent experience for the customer and how that gets supported and diagnosed. Specifically, what it means for us today is that, as we run any integration in our cloud, there is a very consistent set of diagnostics, reporting, metrics, error handling, error tracking that is generated and that's consistent across the many types of integrations that we run.

Again, as our partners become more savvy at working with us, and they know more about that, they can then more consistently offer resolution and support to the customers in the context of the overall Workday support process.

For us, it’s really a way of building this extended and consistent network of support capability and of trust. Where customers have consistent experiences, they have consistent expectations around how and when they get support.

The most frustrating thing is when you are calling one company and they're telling you to call the other company, and there isn’t any consistency or it’s hard to get to the bottom of that. We're hopeful that enlightened integrations around business processes, between collaborating companies, as I have described, will help me to get some of that.

Gardner: It certainly sounds like in the coming years the determining factors of who will be the winner in cloud integration won't be necessarily the one with the biggest, baddest platform -- although that's certainly important. But the one that demonstrates the trust, the SLA response, and maintenance, and generally who becomes a good partner in a diverse and expanding ecosystem will win.

Clarke: That's right. The technology is important, but it's not enough. People just don't just want technology. They want well-intentioned and an honest collaboration between their vendors to help them do the stuff efficiently.

Gardner: Very good. Thanks. You've been listening to a sponsored BriefingsDirect podcast on how major trends around cloud, mobile, and SaaS are dramatically changing requirements and benefits of integration. For more information on Workday's integration as a service, go to http://www.workday.com/solutions/technology/integration_cloud.php.

I would like to thank our guest. We have been here with David Clarke, Director of Integration at Workday. Thanks so much, David.

Clarke: Thanks, 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 and Podcast.com. Download the transcript. Sponsor: Workday.

Transcript of a BriefingsDirect podcast of the role of cloud and SaaS in the changing landscape of application integration. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in:

Friday, July 15, 2011

SaaS PPM from HP Helps Deloitte-Australia Streamline Top-Level Decision Making

Transcript of a sponsored BriefingsDirect podcast, part of a series on application lifecycle management and HP ALM 11.

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

Dana Gardner: Hello, and welcome to a BriefingsDirect podcast on Deloitte-Australia, and how their business has benefited from leveraging the software-as-a-service (SaaS) model for project and portfolio management (PPM) activities.

We spoke to Deloitte-Australia at a recent conference to explore some major enterprise software and solutions, trends and innovations making news across HP’s ecosystem of customers, partners, and developers.

I'm Dana Gardner, Principal Analyst at Interarbor Solutions and I’ll be your host as we learn more about Deloitte’s innovative use of SaaS hosting for non-core applications. First, please join me in welcoming Ferne King, director within the Investment and Growth Forum at Deloitte-Australia in Melbourne. Welcome.

It’s the only solution that we found in the marketplace that would help us support and have visibility into the investments we were going to make.



Ferne King: Thank you.

Gardner: Tell me, Ferne, what led up to the use of SaaS for PPM? Why did this particular model seem to make sense?

King: The SaaS model made sense to us, because we had a strategic direction in our firm that any non-core application’s strategic intent was to have them as SaaS.

Gardner: What is it about PPM, in particular, that made this a leading candidate for that initiative?

King: It’s the only solution that we found in the marketplace that would help us support and have visibility into the investments we were going to make for ourselves internally around the growth and the maintenance of what we did internally within our own firm.

Gardner: Perhaps for our listeners’ benefit, you could explain a little bit about what you do there at Deloitte in Australia and the extent to which you went with this PPM solution. How far and wide?

King: Deloitte-Australia is approximately 5,000 practitioners. In 2010, our revenue was A$850 million. We provide professional services to public and private clients, and we are now globally the largest professional services firm. We utilize PPM internally within the firm, and that helps us to understand that portfolio and prioritization. Deloitte-UK practice and Deloitte-America practice in their consulting areas use PPM to go to the market and help manage deliver investments with their client base.

Gardner: From a high level, what have been some of the benefits of PPM in general, but furthermore, the SaaS deployment method?

Three benefits


King: The three benefits of PPM, primarily for us has been understanding that portfolio and linking that to our strategy. For example, our executive will have a series of business objectives they want to achieve in the Australian practice.

By utilizing PPM, we can understand what is going on within the firm that’s meeting those objectives, and then, more importantly for them, with the gap, and then they can take the action on the gap. That’s the number one priority. The number two priority is being able to communicate to our people within the practice the particulars of change.

For example, over the next quarter, what will our practitioners in the firm see as a result of all of these myriad of initiatives going on, whether it’s a SaaS service HR system or whether it’s a new product and service that they can take to market. Whatever change is coming, we can better communicate that to them within their organization.

Our third priority which the PPM product helps with discipline is our area of delivery. So, in our project management methodology, it helps us improve our disciplines. We had a journey of 18 months of doing things manually and then we brought PPM to technology enable what we were doing manually.

From a SaaS perspective, the benefit we’ve achieved there is that we can focus our people on the right things. Instead of having our people focused on what hardware, what platform, what change request, what design do we need to be happening, we can focus on what our to-be process should be, what our design should be. Then we basically hand that over the fence to the SaaS team, which then help execute that.

We don’t have to stand in a queue within our own IT group and look for a change window. We can make changes every Wednesday, every Sunday, 12 months of the year, and that works for us.

Gardner: Great. Now you described how you’ve decided on the prioritization of moving to SaaS as non-core. But, forgive me, PPM, the way you described it, it sounds rather mission critical. How was it that you decided on this, even though it seems so pertinent and perhaps it forms mortar to other types of business processes?

King: We would say it’s non-core, but high value. Just because it’s non-core, doesn’t mean that it’s not a top priority. Our firm has approximately 2,500 applications within our Australian practice. PPM, at our executive level, is seen as one of our top 10 applications to help our executive, our partners, the senior groups of our firm register ideas to help our business grow and be maintained.

So, it’s high value but it’s not part of our core practice suite. It doesn’t bring in revenue and it doesn’t keep the lights on, but it helps us manage our business.

Gardner: Could you give me a sense of the timeline? Is this something that is relatively new, something that you're well into, something that you're looking at through the rear-view mirror and assessing? Where are we within your roll-out of this PPM for SaaS?

The roadmap


King: I’d answer that in the question of where are we in our roadmap of strategic enterprise portfolio management. In that journey, we're four years in and we are two years into technology enablement. We undertook the journey four years ago to go down strategic portfolio management and we lasted about 18 months to 2 years, manually developing and understanding our methodology, understanding the value where we wanted to go to.

In our second year we technology enabled that to help us execute more effectively, speed to value, time to value, and now we are entering our third year into the maturity model of that.

Gardner: When you describe this as something that doesn’t directly impact your bottom-line, it certainly sounds as if it’s impacting the users within your organization. What’s been the feedback from the front lines, those who are involved with are actually using this? Has this been something that they’ve embraced or something they needed to learn to embrace over time?

King: Fantastic results, particularly at the executive levels who are the ones who pay for us to create the time to work on this. Deloitte itself has taken the transformation over the years, If anybody in the market follows the professional services, industry group areas, Deloitte globally is 160,000 practitioners and over 250 billion of revenue on FY10. We're coming together and have been taking a journey for some time to be as one.

So, if you're a client in the marketplace, you don’t have to think about what door you need to enter the Deloitte world. You enter one door and you get service from whatever service group you need.

PPM has enabled us to help the executive achieve their vision of firm-wide visibility of the enterprise investments we are making to improve our growth and support our maintenance.



If I take the example of three years ago, our tax group would only be interested in what’s happening in their tax group. Our consultant group would really only be interested in what’s happening in the consultant group.

Now that we are acting as one, the tax service line lead and the consulting service line lead would like visibility of what’s happening firm wide. PPM is now enabling us to do that.

What I would summarize there is that PPM has enabled us to help the executive achieve their vision of firm-wide visibility of the enterprise investments we are making to improve our growth and support our maintenance.

Gardner: Tell me if you could, Ferne, a little bit about how you came to the HP solution. What was it that you were looking for in a solution and what requirements did you feel were most important in leading up to that choice?

King: First of all, probably, 27 years experience with project delivery, coming from an engineering construction background, getting very detailed knowledge over the years about the one-on-one delivery components and dealing with a lot of vendors over the years in the client marketplace.

So, well-versed in what we needed and well-versed in what was available out there in the marketplace. When we went to market looking for a partner and a vendor solution, we were very clear on what we wanted. HP was are able to meet that.

I actually took my own role out of the scoring process. We helped put scripts together, scenarios for our vendors to come and demonstrate to us how we were going to achieve meeting our objectives. Then, we brought people around the table from that business with a scoring method, and HP won on that scoring method.

Gardner: With you being fully into this project and for those listening who might be considering moving to a more holistic PPM approach and perhaps evaluating the different sourcing options, is there anything in hindsight that you could offer in terms of advice when beginning such a project?

When we went to market looking for a partner and a vendor solution, we were very clear on what we wanted. HP was are able to meet that.



King: Understand the method or the approach that you want to use PPM for. You cannot bring PPM and expect it to answer 80 percent of your issues. It can support and help direct resolution of issues, but you need to understand how you are expecting to do that. An example would be if you want to capture ideas from other business units or groups or the technology department on what they'd like to do to improve their application or improve product development, any area of the business, understand the life-cycle of how you want that to be managed. Don’t expect PPM to have preset examples for you.

Gardner: Very good. We’ve been discussing how Deloitte-Australia has benefited from leveraging SaaS as a model for delivery of project and portfolio management activities. I’d like to thank our guest, Ferne King, Director within the Investment and Growth Forum at Deloitte-Australia. Thanks so much.

King: Thank you, Dana.

Gardner: And I’d also like to thank our listeners in joining us with this BriefingsDirect podcast, coming to you from a recent HP conference.

I'm 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 and Podcast.com. Download the transcript. Sponsor: HP.

Transcript of a sponsored BriefingsDirect podcast, part of a series on application lifecycle management and HP ALM 11. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in:

Monday, July 11, 2011

Enterprise Architects Increasingly Leverage Advanced TOGAF 9 for Innovation, Market Response, and Governance Benefits

Transcript of a sponsored BriefingsDirect podcast, in conjunction with The Open Group Austin 2011 Conference, on advances in enterprise architecture and The Open Group's Architectural Framework use for business benefits.

Listen to the podcast. Find it on iTunes/iPod and Podcast.com. Download the transcript. Sponsor: The Open Group.

Join The Open Group in Austin, Texas July 18-22 to learn more about enterprise architecture, cloud computing, and TOGAF 9. To register, go to http://www.opengroup.org/austin2011/register.htm.

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 in conjunction with the latest Open Group Conference in Austin, Texas, the week of July 18, 2011. We've assembled a panel to examine the maturing use of The Open Group Architecture Framework (TOGAF), and how enterprise architects and business leaders are advancing and exploiting the latest Version 9.

We'll further explore how the full embrace of TOGAF, its principles, and methodologies are benefiting companies in their pursuit of improved innovation, responsiveness to markets, and operational governance. [Disclosure: The Open Group is a sponsor of BriefingsDirect podcasts.]

Is enterprise architecture (EA) joining other business transformation agents as a part of a larger and extended strategic value? How? And what exactly are the best practitioners of TOGAF getting for their efforts in terms of business achievements?

Here with us to delve into advanced use and expanded benefits of EA frameworks and what that's doing for their user organizations is Chris Forde, Vice President of Enterprise Architecture and Membership Capabilities for The Open Group, based in Shanghai. Welcome, Chris.

Chris Forde: Good morning, Dana.

Gardner: We're also here with Jason Uppal, Chief Architect at QR Systems, based in Toronto. Welcome, Jason.

Jason Uppal: Thank you, Dana.

Gardner: Jason, let’s cut to the quick here. We hear an awful lot about architecture. We hear about planning and methodologies, putting the right frameworks in place, but is TOGAF having an impact on the bottom line in the organizations that are employing it?

Uppal: One of the things with a framework like a TOGAF is that, on the outside, it’s a framework. At the same time, when you apply this along with the other disciplines, it's making big difference in the organization, partially it’s because it's allowing the IT organizations to step up to the plate within the core enterprise as a whole and ask how they can actually exploit the current assets that they already have.

And secondly, [TOGAF helps] make sure the new assets that they do bring into the organization are aligned to the business needs.

One of the examples where EA has a huge impact in many of the organizations that I have experience with is that, with key EA methods, we're able to capture the innovation that exists in the organization and make that innovation real, as opposed to just suggestions that are thrown in a box, and nobody ever sees.

Gardner: What is it about capturing that innovation that gets us to something we can measure in terms of an achievable bottom-line benefit?

Evolve over time

Uppal: Say you define an end-to-end process using architecture development method (ADM) methods in TOGAF. What it does is give me a way to capture that innovation at the lowest level and then evolve it over time. Those people who are part of the innovation at the beginning see their innovation or idea progressing through the organization, as the innovation gets aligned to value statements, and value statements get aligned to their capabilities, and the strategies, and the projects, and hence to the end of the day.

Therefore, if I make a suggestion of some sort, that innovation or idea is seen throughout the organization through the methods like ADM, and the linkage is explicit and very visible to the people. Therefore, they feel comfortable that their ideas are going somewhere, they are just not getting stuck.

Forde: There's an additional point here, Dana, to underscore the answer that Jason gave to your question. In the end result, what you want to be seeing out of your architectural program is moving the key performance indicators (KPIs) for the business, the business levers that they are expecting to be moved out. If that is related to cost reduction or is related to top-line numbers or whatever, that explicit linkage through to the business levers in an architecture program is critical.

Gardner: Chris, you have a good view on the global markets and the variability of goals here. Many companies are looking to either cut cost or improve productivity. Others are looking to expand. Others are looking to manage how to keep operations afloat. There are so many different variables. How do things like the TOGAF 9 and EA have a common benefit to all of those various pursuits? What is the common denominator that makes EA so powerful?

Forde: Going back to the framework reference, what we have with TOGAF 9 is a number of assets, but primarily it’s a tool that’s available to be customized, and it's expected to be customized.

If you come to the toolset with a problem, you need to focus the framework on the area that's going to help you get rapid value to solving your particular problem set. So once you get into that particular space, then you can look at migrating out from that entry point, if that's the approach, to expanding your use of the framework, the methods, the capabilities, that are implicit and explicit in the framework to address other areas.

You can start at the top and work your way down through the framework, from this kind of ├╝ber value proposition, right down through delivery to the departmental level or whatever. Or, you can come into the bottom, in the infrastructure layer, in IT for example, and work your way up. Or, you can come in at the middle. The question is what is impeding your company’s growth or your department’s growth, if those are the issues that are facing you.

One of the reasons that this framework is so useful in so many different dimensions is that it is a framework. It’s designed to be customized, and is applicable to many different problems.

Gardner: Back to you, Jason. When we think about a beginning effort, perhaps a crawl-walk-run approach to EA and TOGAF, the promise is that further development, advancement, understanding, implementation will lead to larger, more strategic goals.

Let’s define what it means to get to that level of maturity. When we think about an advanced user of TOGAF, what does that mean? Then, we'll get into how they can then leverage that to further goals. But, what do we really mean by an advanced user at this point?

Advanced user

Uppal: When we think about an advanced user, in our practice we look at it from different points of view and ask what value I'm delivering to the organization. It could very well be delivering value to a CTO in the organization. That is not to say that's not an advanced user, because that’s strictly focused on technology.

But then, the CTO focus is that it allows us to focus on the current assets that are under deployment in the organization. How do you get the most out of them? So, that’s an advanced user who can figure out how to standardize and scale those assets into a scalable way so therefore they become reusable in the organization.

As we move up the food chain from very technology-centric view of a more optimized and transformed scale, advanced user at that point is looking at and saying that now I have a framework like TOGAF, that advanced user has all these tools in their back pocket.

Now, depending on the stakeholder that they're working with, be that a CEO, a CFO, or a junior manager in the line of business, I can actually focus them on defining a specific capability that they are working towards and create transition roadmaps. Once those transition roadmaps are established, then I can drive that through.

An advanced user in the organization is somebody who has all these tools available to them, frameworks available to them, but at the same time, are very focused on a specific value delivery point in their scope.

It moves the conversation away from this framework debate and very quickly moves our conversation into what we do with it.



One beauty of TOGAF is that, because we get to define what enterprise is and we are not told that we have to interview the CEO on day one, I can define an enterprise from a manager’s point of view or a CFO’s point of view and work within that framework. That to me is an advanced user.

Gardner: When we talk about applied architecture, what does that mean? How is it that we move from concept into execution?

Uppal: The frameworks that we have are well thought-out frameworks. So, it moves the conversation away from this framework debate and very quickly moves our conversation into what we do with it.

When we talk about a framework like TOGAF, now I can look at and say that if I wanted to apply it now, I have an executive who has defined a business strategy, which typically is a two page PowerPoint presentation, sometimes accompanied by Excel. That’s a good starting point for an enterprise architect.

Now, I use methods like TOGAF to define the capabilities in that strategy that they are trying to optimize, where they are, and what they want to transition to.

Very creative

This is where a framework allows me to be very creative, defining the capabilities and the transition points, and giving a roadmap to get to those transitions. That is the cleverness and cuteness of architecture work, and the real skills of an architect comes into, not in defining the framework, but defining the application of the framework to a specific business strategy.

Gardner: Jason, we mentioned that there is a great deal of variability in what different companies in different regions and in different industries need to accomplish, but one of the common questions I get a lot these days is what to outsource and what to manage internally and how to decide the boundaries between a core competency and extended outsourcing or hybrid computing types of models?

How does the applied architecture come to the rescue, when this sort of question, which I think is fundamental to an enterprise, comes up?

Uppal: That’s a great question. That’s one of the area where if architects do their job well, we can help the organization move much further along. Because, what we do in the business space, and we have done it many times with the framework, is to look at the value chain of the organization. And looking at the value chain, then to map that out to the capabilities required.

Once we know those capabilities, then I can squarely put that question to the executives and say, "Tell me which capability you want to be the best at. Tell me what capability you want to lead the market in. And, tell me which capability you want to be mediocre and just be at below the benchmark in industry."

Once I get an understanding of which capability I want to be the best at, that's where I want to focus my energy.



Once I get an understanding of which capability I want to be the best at, that's where I want to focus my energy. Those ones that I am prepared to live with being mediocre, then I can put another strategy into place and ask how I outsource these things, and focus my outsourcing deal on the cost and service.

This is opposed to having very confused contract with the outsourcer, where one day I'm outsourcing for the cost reasons. The other day, I'm outsourcing for growth reasons. It becomes very difficult for an organization to manage the contracts and bend it to provide the support.

That conversation, at the beginning, is getting executives to commit to which capability they want to be best at. That is a good conversation for an enterprise architect.

My personal experience has been that if I get a call back from the executive, and they say they want to be best at every one of them, then I say, "Well, you really don’t have a clue what you are talking about. You can’t be super fast and super good at every single thing that you do."

Gardner: So making those choices is what’s critical. Some of the confusion I also hear about in the field is how to do a cost-benefit analysis about what processes I might keep internal, versus either hybrid or external source processes?

Is there something about the applied architecture and TOGAF 9 that sets up some system of record or methodology approach that allows that cost-benefit analysis of these situations to be made in advance? Is there anything that the planning process brings to the table in trying to make proper decisions about sourcing?

Capability-based planning

Uppal: Absolutely. This is where the whole of our capability-based planning conversation is. It was introduced in TOGAF 9, and we got more legs to go into developing that concept further, as we learn how best to do some of these things.

When I look at a capability-based planning, I expect my executives to look at it from a point of view and ask what are the opportunities and threats. What it is that you can get out there in the industry, if you have this capability in your back pocket? Don’t worry about how we are going to get it first, let’s decide that it’s worth getting it.

Then, we focus the organization into the long haul and say, well, if we don’t have this capability and nobody in the industry has this capability, if we do have it, what will it do for us? It provides us another view, a long-term view, of the organization. How are we going to focus our attention on the capabilities?

One of the beauties of doing EA is, is that when we start EA at the starting point of a strategic intent, that gives us a good 10-15 year view of what our business is going to be like. When we start architecture at the business strategy level, that gives us a six months to five-year view.

Enterprise architects are very effective at having two views of the world -- a 5-, 10-, or 15-year view of the world, and a 6-month to 3-year view of the world. If we don’t focus on the strategic intent, we'll never know what is possible, and we would always be working on what is possible within our organization, as opposed to thinking of what is possible in the industry as a whole.

Everybody is trying to understand what it is they need to be good at and what it is their partners are very good at that they can leverage.



Gardner: So, in a sense, you have multiple future tracks or trajectories that you can evaluate, but without a framework, without an architectural approach, you would never be able to have that set of choices before you.

Chris Forde, any thoughts on what Jason’s been saying in terms of the sourcing and cost benefits and risks analysis that go into that?

Forde: In the kinds of environment that most organizations are operating in -- government, for-profit, not-for-profit organizations -- everybody is trying to understand what it is they need to be good at and what it is their partners are very good at that they can leverage. Their choices around this are of course critical.

One of the things that you need to consider is that if you are going to give x out and have the power to manage that and operate whatever it is, whatever process it might be, what do you have to be good at in order to make them effective? One of the things you need to be good at is managing third parties.

One of the advanced uses of an EA is applying the architecture to those management processes. In the maturity of things you can see potentially an effective organization managing a number of partners through an architected approach to things. So when we talked about what do advanced users do, what I am offering is that an advanced use of EA is in the application of it to third-party management.

Gardner: So the emphasis is on the process, not necessarily who is executing on that process?

Framework necessary

Forde: Correct, because you need a framework. Think about what most major Fortune 500 companies in the United States do. They have multiple, multiple IT partners for application development and potentially for operations. They split the network out. They split the desktop out. This creates an amazing degree of complexity around multiple contracts. If you have an integrator, that’s great, but how do you manage the integrator?

There’s a whole slew of complex problems. What we've learned over the years is that the original idea of “outsourcing,” or whatever the term that’s going to be used, we tend to think of that in the abstract, as one activity, when in fact it might be anywhere from 5-25 partners. Coordinating that complexity is a major issue for organizations, and taking an architected approach to that problem is an advanced use of EA.

Join The Open Group in Austin, Texas July 18-22 to learn more about enterprise architecture, cloud computing, and TOGAF 9. To register, go to http://www.opengroup.org/austin2011/register.htm.

Gardner: So stated another way, Jason, the process is important, but the management of processes is perhaps your most important core competency. Is that fair, and how does EA support that need for a core competency of managing processes across multiple organizations?

Uppal: That’s absolutely correct. Chris is right. For example, there are two capabilities an organization decided on, one that they wanted to be very, very good at.

We worked with a large concrete manufacturing company in the northern part of the country. If you're a concrete manufacturing company, your biggest cost is the cement. If you can exploit your capability to optimize the cement and substitute products with the chemicals and get the same performance, you can actually get a lot more return and higher margins for the same concrete.

The next thing is the cleverness of the architect -- how he uses his tools to actually define the best possible solutions.



In this organization, the concrete manufacturing process itself was core competency. That had to be kept in-house. The infrastructure is essential to make the concrete, but it wasn’t the core competency of the organization. So those things had to be outsourced.

In this organization we have to build a process -- how to manage the outsourcer and, at the same time, have a capability and a process. Also, how to become best concrete manufacturers. Those two essential capabilities were identified.

An EA framework like TOGAF actually allows you to build both of those capabilities, because it doesn’t care. It just thinks, okay, I have a capability to build, and I am going to give you a set of instructions, the way you do it. The next thing is the cleverness of the architect -- how he uses his tools to actually define the best possible solutions.

Gardner: Of course, it’s not just enough to identify and implement myriad sourcing or complex sourcing activities, but you need to monitor and have an operational governance oversight capability as well. Is there something in TOGAF 9 specifically that lends itself to taking this into the operational and then creating ongoing efficiencies as a result?

Uppal: Absolutely, because this is one of the areas where in ADM, when we get back to our implementation of governance, and post implementation of governance, value realization, how do we actually manage the architecture over the life of it? This is one of the areas where TOGAF 9 has done a considerably good job, and we've still got a long way to go in how we actually monitor and what value is being realized.

Very explicit

Our governance model is very explicit about who does what and when and how you monitor it. We extended this conversation using TOGAF 9 many times. At the end, when the capability is deployed, the initial value statement that was created in the business architecture is given back to the executive who asked for that capability.

We say, "This is what the benefits of these capabilities are and you signed off at the beginning. Now, you're going to find out that you got the capability. We are going to pass this thing into strategic planning next year, because for next year's planning starting point, this is going to be your baseline." So not only is the governance just to make sure it’s via monitoring, but did we actually get the business scores that we anticipated out of it.

Gardner: Another area that’s of great interest to me nowadays is looking at the IT organization as they pursue things like cloud, software as a service (SaaS), and hybrid models. Do they gather a core competency at how to manage these multiple partners, as Chris pointed out, or does another part of the company that may have been dealing with outsourcing at a business process level teach the IT department how to do this?

Any sense from either of our panelists on whether IT becomes a leader or a laggard in how to manage these relationships, and how important is managing the IT element of that in the long run? Let’s start with you, Jason.

Uppal: It depends on the industry the IT is in. For example, if you're an organization that is very engineering focused, engineers have a lot more experience managing outsourcing deals than IT organizations do. In that case, the engineering leads this conversation.

We have actually done cloud before. This is not a new thing, except that before we looked at it from a hosting point of view and from a SaaS point of view.



But in most organizations, which are service-oriented organizations, engineering has not been a primary discipline, and IT has a lot of experience managing outside contracts. In that case, the whole cloud conversation becomes a very effective conversation within the IT organization.

When we think about cloud, we have actually done cloud before. This is not a new thing, except that before we looked at it from a hosting point of view and from a SaaS point of view. Now, cloud is going in a much further extended way, where entire capability is provided to you. That capability is not only that the infrastructure is being used for somebody else, but the entire industry’s knowledge is in that capability.

This is becoming a very popular thing, and rightfully so, not because it’s a sexy thing to have. In healthcare, especially in countries where it’s a socialized healthcare and it's not monopolized, they are sharing this knowledge in the cloud space with all the hospitals. It's becoming a very productive thing, and enterprise architects are driving it, because we're thinking of capabilities, not components.

Gardner: Chris Forde, similar question. How do you see the role of IT shifting or changing as a result of the need to manage more processes across multiple sources?

Forde: It’s an interesting question. I tend to agree with the earlier part of Jason’s response. I am not disagreeing with any of it, actually, but the point that he made about it is that it's a "it depends" answer.

IT interaction

Under normal circumstances the IT organizations are very good at interacting with other technology areas of the business. From what I've seen with the organizations I have dealt with, typically they see slices of business processes, rather than the end-to-end process entirely.

Even within the IT organizations typically, because of the size of many organizations, you have some sort of division of responsibilities. As far as Jason’s emphasis on capabilities and business processes, of course the capabilities and processes transcend functional areas in an organization.

To the extent that a business unit or a business area has a process owner end to end, they may well be better positioned to manage the BPM outsourcing-type of things. If there's a heavy technology orientation around the process outsourcing, then you will see the IT organization being involved to one extent or another.

The real question is, where is the most effective knowledge, skill, and experience around managing these outsourcing capabilities? It may be in the IT organization or it may be in the business unit, but you have to assess where that is.

Under normal circumstances the IT organizations are very good at interacting with other technology areas of the business.



That's one of the functions that the architecture approaches. You need to assess what it is that's going to make you successful in this. If what you need happens to be in the IT organization, then go with that ability. If it is more effective in the business unit, then go with that. And perhaps the answer is that you need to combine or create a new functional organization for the specific purpose of meeting that activity and outsource need.

I'm hedging a little bit, Dana, in saying that it depends.

Gardner: It certainly raises some very interesting issues. At the same time that we're seeing this big question mark around sourcing and how to do that well, we're also in a period where more organizations are being data-driven and looking to have deeper, more accessible, and real-time analytics applied to their business decisions. Just as with sourcing, IT also has an integral role in this, having been perhaps the architects or implementators of warehousing, data marts, and business intelligence (BI).

Back to you Jason. As we enter into a phase where organizations are also trying to measure and get scientific and data-driven about their decisions, how does IT, and more importantly, how does TOGAF and EA come to help them do that?

Uppal: We have a number of experiences like that, Dana. One is a financial services organization. The entire organization’s function is that they manage some 100-plus billion dollars worth of assets. In that kind of organization, all the decision making process is based on the data that they get. And 95 percent of the data is not within the organization. It is vendor data that they're getting from outside.

So in that kind of conversation, we look and say that the organization needs a capability to manage data. Once we define a capability, then we start putting metrics on this thing. What does this capability need to be able to do?

In this particular example, we put a metric on this and said that the data gets identified in the morning, by the afternoon we bring it into the organization, and by the end of the day we get rid of it. That’s how fast the data has to be procured, transformed into the organization, brought it in, and delivered it to end-use. That end-user makes the decision whether we will never look at the data again.

Data capability

H
aving that fast speed of data management capability in the organization, and this is one of the areas where architects can take a look at, this is the capability you need. Now I can give you a roadmap to get to that capability.

Gardner: Chris Forde, how do you see the need for a data-driven enterprise coincide with IT and EA?

Forde: For most, if not all, companies, information and data are critical to their operation and planning activities, both on a day-to-day basis, month-to-month, annually, and in longer time spans. So the information needs of a company are absolutely critical in any architected approach to solutions or value-add type of activities.

I don’t think I would accept the assumption that the IT department is best-placed to understand what those information needs are. The IT organization may be well-placed to provide input into what technologies could be applied to those problems, but if the information needs are normally being applied to business problems, as opposed to technology problems, I would suggest that it is probably the business units that are best-placed to decide what their information needs are and how best to apply them.

The technologist’s role, at least in the model I'm suggesting, is to be supportive in that and deliver the right technology, at the right time, for the right purpose.

Gardner: Then, how would a well-advanced applied architecture methodology and framework help those business units attain their information needs, but also be in a position to exploit IT’s helping hand when needed?

We have good enough frameworks. We're moving to the next phase of what we do with these frameworks.



Forde: It's mostly providing the context to frame the problem in a way that it can be addressed, chunked down to reasonable delivery timeframes, and then marshaling the resources to bring that to reality.

From a pure framework and applied methodology standpoint, if you're coming at it from an idealized situation, you're going to be doing it from a strategic business need and you're going to be talking to the business units about what their capability and functional needs are. And at that time, you're really in the place of what business processes they're dealing with and what information they need in order to accomplish what the particular set of goals is.

This is way in advance of any particular technology choice being made. That's the idealized situation, but that’s typically what most frameworks, and in particular, the TOGAF 9 Framework from The Open Group, would go for.

Gardner: We're just beginning these conversations about advanced concepts in EA and there are going to be quite a bit more offerings and feedback and collaboration around this subject at The Open Group Conference in Austin. Perhaps before we sign off, Jason, you can give us a quick encapsulation of what you will be discussing in terms of your presentation at the conference.

Uppal: One of the things that we've been looking at from the industry’s point of view is saying that this conversation around the frameworks is a done deal now, because everybody accepted that we have good enough frameworks. We're moving to the next phase of what we do with these frameworks.

In our future conferences, we're going to be addressing that and saying what people are specifically doing with these frameworks, not to debate the framework itself, but the application of it.

Continuous planning

I
n Austin we'll be looking at how we're using a TOGAF framework to improve ongoing annual business and IT planning. We have a specific example that we are going to bring out where we looked at an organization that was doing once-a-year planning. That was not a very effective way for the organizations. They wanted to change it to continuous planning, which means planning that happens throughout the year.

We identified four or five very specific measurable goals that the program had, such as accuracy of your plan, business goals being achieved by the plan, time and cost to manage and govern the plan, and stakeholders’ satisfaction. Those are the areas that we are defining as to how the TOGAF like framework will be applied to solve a specific problem like enterprise planning and governance.

That's something we will be bringing to our conference in Austin and that event will be held on a Sunday. In the future, we'll be doing a lot more of those specific applications of a framework like a TOGAF to a unique set of problems that are very tangible and they very quickly resonate with the executives, not in IT, but in the entire organization.

Forde: Can I follow along with a little bit of a plug here, Dana.

Gardner: Certainly.

Forde: Jason is going to be talking as a senior architect on the applied side of TOGAF on this Sunday. For the Monday plenary, this is basically the rundown. We have David Baker, a Principal from PricewaterhouseCoopers, talking about business driven architecture for strategic transformations.

This is a time now for the enterprise architects to really step up to the plate and be accountable for real performance influence on the organization’s bottom line.



Following that, Tim Barnes, the Chief Architect at Devon Energy out of Canada, covering what they are doing from an EA perspective with their organization.

Then, we're going to wrap up the morning with Mike Wolf, the Principal Architect for EA Strategy and Architecture at Microsoft, talking about IT Architecture to the Enterprise Architecture.

This is a very powerful lineup of people addressing this business focus in EA and the application of it for strategic transformations, which I think are issues that many, many organizations are struggling with.

Gardner: Looking at, again, the question I started us off with, how do TOGAF and EA affect the bottom line? We've heard about how it affects the implementation for business transformation processes. We've talked about operational governance. We looked at how sourcing, business process management and implementation, and ongoing refinement are impacted. We also got into data and how analytics and information sharing are affected. Then, as Jason just mentioned, planning and strategy as a core function across a variety of different types of business problems.

So, I don’t think we can in any way say that there's a minor impact on the bottom line from this. Last word to you, Jason.

Uppal: This is a time now for the enterprise architects to really step up to the plate and be accountable for real performance influence on the organization’s bottom line.

If we can improve things like exploiting assets better today than what we have, improve our planning program, and have very measurable and unambiguous performance indicator that we're committing to, this is a huge step forward for enterprise architects and moving away from technology and frameworks to real-time problems that resonate with executives and align to business and in IT.

Gardner: Well, great. You've been listening to a sponsored podcast discussion in conjunction with The Open Group Conference in Austin, Texas, the week of July 18, 2011.

I would like to thank our guests. We have been joined by Chris Forde, Vice President of Enterprise Architecture and Membership Capabilities for The Open Group. Thanks, Chris.

Forde: Thanks, Dana.

Gardner: And also Jason Uppal, Chief Architect at QR Systems. Thank you, Jason.

Uppal: Thank you, Dana.

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

Listen to the podcast. Find it on iTunes/iPod and Podcast.com. Download the transcript. Sponsor: The Open Group.

Transcript of a sponsored BriefingsDirect podcast, in conjunction with The Open Group Austin 2011 Conference, on advances in enterprise architecture and The Open Group's Architectural Framework use for business benefits. Copyright The Open Group and Interarbor Solutions, LLC, 2005-2011. All rights reserved.

Join The Open Group in Austin, Texas July 18-22 to learn more about enterprise architecture, cloud computing, and TOGAF 9. To register, go to http://www.opengroup.org/austin2011/register.htm.

You may also be interested in:

Wednesday, July 06, 2011

Case Study: T-Mobile's Massive Data Center Transformation Journey Wins Award Using HP ALM Tools

Transcript of a BriefingsDirect podcast on how awarding-winning communications company T-Mobile improved application quality, while setting up two new data centers.

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

Dana Gardner: Hello, and welcome to a special BriefingsDirect podcast series coming to you from the HP Discover 2011 conference in Las Vegas. We're here on the Discover show floor the week of June 6 to explore some major enterprise IT solution trends and innovations making news across HP’s ecosystem of customers, partners, and developers.

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

Our latest user case study focuses on an award-wining migration and transformation and a grand-scale data center transition for T-Mobile. I was really impressed with the scope and size and the amount of time -- in terms of being short -- for you all to do this.

We're here with two folks who are going to tell us more about what T-Mobile has done to set up two data centers, and how in the process they have improved their application quality and the processes behind their application lifecycle management (ALM). [Disclosure: HP is a sponsor of BriefingsDirect podcasts.]

So join me in welcoming Michael Cooper, Senior Director of Enterprise IT Quality Assurance at T-Mobile. Welcome, Michael.

Michael Cooper: Thank you.

Gardner: We're also here with Kirthy Chennaian, Director Enterprise IT Quality Management at T-Mobile. Welcome.

Kirthy Chennaian: Thank you. It's a pleasure.

Gardner: People don’t just do these sorts of massive, hundred million dollar-plus activities because it's nice to have. This must have been something that was really essential for you.

Cooper: Absolutely. There are some definite business drivers behind setting up a world-class, green data center and then a separate disaster-recovery data center. Just for a little bit of a clarification. The award that we won is primarily focused on the testing effort and the quality assurance (QA) effort that went into that.

Gardner: Kirthy, tell me why you decided to undertake both an application transformation as well as a data center transformation -- almost simultaneously?

Chennaian: Given the scope and complexity of the initiative, ensuring system availability was primarily the major driver behind this. QA plays a significant role in ensuring that both data centers were migrated simultaneously, that the applications were available in real-time, and that from a quality assurance and testing standpoint we had to meet time-frames and timelines.

Gardner: Let's get a sense of the scope. First and foremost, Michael, tell me about T-Mobile and its stature nowadays.

Significant company

Cooper: T-Mobile is a national provider of voice, data, and messaging services. Right now, we're the fourth largest carrier in the US and have about 33 million customers and $21 billion in revenue, actually a little bit more than that. So, it's a significant company.

We're a company that’s really focused on our customers, and we've gone through an IT modernization. The data center efforts were a big part of that IT modernization, in addition to modernizing our application platform.

Gardner: Let's also talk about the scope of your movement to a new data center, and then we can get into the application transformation parts of that. In a nutshell, what did we do here? It sounds like we've set up two modern data centers, and then migrated your apps and data from an older one into those.

Chennaian: Two world-class data centers, as Michael had pointed out. One in Wenatchee, Washington and the other one is Tempe, Arizona. The primary data center is the one in Wenatchee, and the failover disaster-recovery data center is in Tempe, Arizona.

Cooper: What we were doing was migrating more than 175 Tier 1 applications and Tier 0, and some Tier 2 as well. It was a significant effort requiring quite a bit of planning, and the HP tools had a big part in that, especially in the QA realm.

Gardner: Now, were these customer-facing apps, internal apps, logistics? Are we talking about retail? Give me a sense of the scope here on the breadth and depth of your apps?

Chennaian: Significant. We're talking critical applications that are customer-facing. We're talking enterprise applications that span across the entire organization. And, we're also talking about applications that support these critical front-end applications. So, as Michael pointed out, 175 applications needed to be migrated across both of the data centers.

For example, moving T-Mobile.com, which is a customer-facing critical application, ensuring that it was transitioned seamlessly and was available to the customer in real-time was probably one of the key examples of the criticality behind ensuring QA for this effort.

Gardner: IT is critical for almost all companies nowadays, but I can't imagine a company where technology is more essential and critical than T-Mobile as a data and services carrier.

What's the case with the customer response? Do you have any business metrics, now that you’ve gone through this, that demonstrate not just that you're able to get better efficiency and your employees are getting better response times from their apps and data, but is there like a tangible business benefit, Michael?

Near-perfect availability

Cooper: I can't give you the exact specifics, but we've had significant increases in our system up-time and almost near-perfect availability in most areas. That’s been the biggest thing.

Kirthy mentioned T-Mobile.com. That’s an example where, instead of the primary and the backup, we actually have an active-active situation in the data center. So, if one goes down the other one is there, and this is significant.

A significant part of the way that we used HP tools in this process was not only the functional testing with Quick Test Professional and Quality Center, but we also did the performance testing with Performance Center and found some very significant issues that would have gone on to production.

This is a unique situation, because we actually got to do the performance testing live in the performance environments. We had to scale up to real performance types of loads and found some real issues that -- instead of the customers facing them, they didn’t have to face them.

The other thing that we did that was unique was high-availability testing. We tested each server to make sure that if one went down, the other ones were stable and could support our customers.

We were able to deliver application availability, ensure a timeframe for the migration and leverage the ability to use automation tools.



Gardner: Now, this was a case where not only were you migrating apps, but you were able to go in and make sure that they were going to perform well within this in new environment. As you pointed out, Michael, you were able to find some issues in those apps in the transition, and at the same time simultaneously you upgraded to the more recent refreshes of the HP products to do that.

So, this was literally changing the wings on the airplane when it was still flying. Tell me why doing it all at once was a good thing.

Chennaian: It was the fact that we were able to leverage the additional functionality that the HP suite of products provide. We were able to deliver application availability, ensure a time-frame for the migration and leverage the ability to use automation tools that HP provides. With Quick Test Professional, for example, we migrated from version 9.5 to 10.0, and we were able to leverage the functionality with business process testing from a Quality Center standpoint.

As a whole, from an application lifecycle management and from an enterprise-wide QA and testing standpoint, it allowed us to ensure system availability and QA on a timely basis. So, it made sense to upgrade as we were undergoing this transformation.

Cooper: Good point, Kirthy. In addition to upgrading our tools and so forth, we also upgraded many of the servers to some of the latest Itanium technology. We also implemented a lot of the state-of-the-art virtualization services offered by HP, and some of the other partners as well.

Streamlined process

Using HP tools, we were able to create a regression test set for each of our Tier 1 applications in a standard way and a performance test for each one of the applications. So, we were able to streamline our whole QA process as a side-benefit of the data migration, building out these state-of-the-art data centers, and IT modernization.

Gardner: So, this really affected operations. You changed some platforms, you adopted the higher levels of virtualization, you're injecting quality into your apps, and you're moving them into an entirely new facility. That's very impressive, but it's not just me being impressed. You've won a People's Choice Award, voted by peers of the HP software community and their Customer Advisory Board. That must have felt pretty good.

Cooper: It feels excellent. In 2009, we won the IT Transformation Award. So, this isn't our first time to the party. That was for a different project. I think that in the community people know who we are and what we're capable of. It's really an honor that the people who are our peers, who read over the different submissions, decided that we were the ones that were at the top.

Gardner: And I hear that you've won some other awards as well.

Cooper: We've won lots of awards, but that's not what we do it for. The reason why we do the awards is for the team. It's a big morale builder for the team. Everybody is working hard. Some of these project people work night and day to get them done, and the proof of the pudding is the recognition by the industry.

Our CIO has a high belief in quality and really supports us in doing this. It's nice that we've got the industry recognition as well.



Honestly, we also couldn't do without great executive support. Our CIO has a high belief in quality and really supports us in doing this. It's nice that we've got the industry recognition as well.

Gardner: Of course, the proof of the pudding is in the eating. You've got some metrics here. They were pretty impressive in turns of availability, cost savings, reduction in execution time, performance and stability improvements, and higher systems availability.

Give me a sense, from an IT perspective, if you were to go to some other organization, not in the carrier business, of course, and tell them what this really did for you, performance and in the metrics that count to IT, what would you tell them?

Cooper: The metrics I can speak to are from the QA perspective. We were able to do the testing and we never missed one of the testing deadlines. We cut our testing time using HP tools by about 50 percent through automation, and we can pretty accurately measure that. We probably have about 30 percent savings in the testing, but the best part of it is the availability. But, because of the sensitive nature and competitive marketplace, we're not going to talk exactly about what our availability is.

Gardner: And how about your particular point of pride on this one, Kirthy?

Chennaian: For one, being able to get recognized is an acknowledgement of all the work you do, and for your organization as a whole. Mike rightly pointed out that it boosts the morale of the organization. It also enables you to perform at a higher level. So, it's definitely a significant acknowledgment, and I'm very excited that we actually won the People's Choice Award.

Gardner: A number of other organizations and other series of industries are going to be facing the same kind of a situation, where it's not just going to be a slow, iterative improvement process,. They're going to have to go catalytic and make wholesale changes in the data center, looking for that efficiency benefit.

You've done that. You've improved on your QA and applications lifecycle benefits at the same time. With that 20-20 hindsight, what would you have done differently, or at least what could you advise people who are going to face a similar large, complex, and multifaceted undertaking?

Planning and strategy

Chennaian: If I were to do this again, I think there is definitely a significant opportunity with respect to planning and investing in the overall strategy of QA and testing for such a significant transformation. There has to be a standard methodology. You have to have the right toolsets in place. You have to plan for the entire transformation as a whole. Those are significant elements in successful transformation.

Gardner: Monday morning quarterback for you, Michael?

Cooper: We did a lot of things right. One of the things that we did right was to augment our team. We didn’t try to do the ongoing work with the exact same team. We brought in some extra specialists to work with us or to back-fill in some places. Other groups didn’t and paid the price, but that part worked out for us.

Also, it helped to have a seat at the table and say, "It's great to do a technology upgrade, but unless we really have the customer point of view and focus on the quality, you're not going to have success."

We were lucky enough to have that executive support and the seat at the table, to really have the go/no-go decisions. I don't think we really missed one in terms of ones that we said, "We shouldn't do it this time. Let's do it next time." Or, ones where we said, "Let's go." I can't remember even one application we had to roll back. Overall, it was very good. The other thing is, work with the right tools and the right partners.

Gardner: With data center transformation, after all, it's all about the apps. You were able to maintain that focus. You didn’t lose focus of the apps?

It's great to do a technology upgrade, but unless we really have the customer point of view and focus on the quality, you're not going to have success.



Cooper: Definitely.The applications do a couple of things. One, the ones that support the customers directly. Those have to have really high availability, and we're able to speed them up quite a bit with the newest and the latest hardware.

The other part are the apps that people don't think about that much, which are the ones that support the front lines, the ones that support retail and customer care and so forth. I would say that our business customers or internal customers have also really benefited from this project.

Gardner: Well great. We've been talking about a massive undertaking with data center transformation and application QA and lifecycle improvements and the result was a People's Choice Award won here at the Discover Show in Las Vegas. It's T-Mobile, the winner. We've been talking with their representatives here. Michael Cooper, the Senior Director of Enterprise IT Quality Assurance. Thanks again, Michael.

Cooper: Thank you, and we're very proud of the team.

Gardner: We are also here with Kirthy Chennaian, the Director of Enterprise IT Quality Management at T-Mobile. Thanks.

Chennaian: Thank you. Very excited to be here.

Gardner: And thanks to our audience for joining this special BriefingsDirect podcast coming to you from the HP Discover 2011 Conference. I'm Dana Gardner, Principal Analyst at Interarbor Solutions, your host for this series of User Experience Discussions. Thanks again for listening, and come back next time.

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

Transcript of a BriefingsDirect podcast on how awarding-winning communications company T-Mobile improved application quality, while setting up two data centers. Copyright Interarbor Solutions, LLC, 2005-2011. All rights reserved.

You may also be interested in: