Tuesday, December 19, 2006

Transcript of BriefingsDirect SOA Insights Edition Vol. 6 Podcast: Year in Review, Predictions for 2007

Edited transcript of weekly BriefingsDirect[TM] SOA Insights Edition, recorded Dec. 1, 2006.

Listen to the podcast here. If you'd like to learn more about BriefingsDirect B2B informational podcasts, or to become a sponsor of this or other B2B podcasts, contact Dana Gardner at 603-528-2435.

Dana Gardner: Hello, and welcome to the latest BriefingsDirect SOA Insights Edition, Volume 6. This is a weekly discussion and dissection of Services Oriented Architecture (SOA) related news and events with a panel of independent IT industry analysts. I’m your host and moderator, Dana Gardner, principal analyst at Interarbor Solutions.

Joining us this week, our panel of independent IT analysts includes show regular Steve Garone. Steve is an independent analyst, a former program vice president at IDC, and the founder of the AlignIT Group. Welcome Steve.

Steve Garone: Hi Dana. It’s great to be here today.

Gardner: Also joining us is Joe McKendrick. He’s an independent research consultant and columnist at Database Trends, as well as a blogger for ZDNet and ebizQ. Welcome, Joe.

Joe McKendrick: Hello, Dana.

Gardner: Also, a return visit from Jon Collins. He’s the principal analyst at Macehiter Ward-Dutton in the UK. Welcome back, Jon.

Jon Collins: Hi there, everyone.

Gardner: Making his debut on the show this week is Tony Baer, principal at onStrategies.

Tony Baer: Hey, Dana.

Gardner: Thanks, Tony, for coming along. This week, the week of Nov. 25, 2006, we’re coming up to the end of the year, and so we thought it would make sense to take some stock of the year in review around SOA issues. We’ve seen an awful lot this year, and 2006 is probably the most prominent year for SOA.

We’ve had a bevy of consolidation events, mergers, and acquisitions. We’ve seen some large vendors further and more deeply embrace SOA. We’ve seen startups realign themselves around the SOA message and value, and we’ve seen SOA extend, perhaps a little bit high in the stack, around application and services, and the application-as-services in more than the infrastructure.

To help us weed through some of this, our panel will be coming up with some of their own insights, as well as a round-robin approach to SOA. Let me start off. From my perspective, the biggest story of the year is consolidation: A lot of purchasing, mergers, and acquisitions. Perhaps the most impactful of these included the Red Hat acquisition of JBoss, as well as the HP acquisition of Mercury, which came on the heels of Mercury’s acquisition of Systinet.

We’ve seen many others. There was Progress and Actional/Sonic, BEA-Flashline, webMethods-Infravio, we’ve seen FileNet go to IBM, and IBM with a slew of other smaller companies. But I want to go around the table and see if there’s some concurrence around this notion that the Red Hat JBoss and the HP Mercury had perhaps the most impact in the field? Why don’t we start with our new analyst this week, Tony?

Baer: Well, I probably wouldn’t disagree too heavily with that. A problem is that it’s becoming increasingly difficult to distinguish software from services, at least in terms of vendor products. So, you almost cast almost any major product introduction or acquisition with an SOA angle or SOA spin to it.

A really great example was Mercury, which prior to the Systinet acquisition, surprisingly didn’t have that big a play in services per se. They were more heavily invested obviously in testing and IT governance. The obvious question is when was this going to extend to SOA governance.

They acquired Systinet, but my sense with Mercury is that they were obviously a bit distracted this year and did not get around to absorbing Systinet as fast as they otherwise could have. Now that it’s under HP’s watch -- and without the distractions of SEC-related inquiries -- hopefully this will go along a little bit faster, and they’ll start to coalesce their SOA strategy.

Gardner: Perhaps we shouldn’t think of these as SOA mergers and acquisitions. It’s just software and services. Maybe the SOA moniker isn’t as important and has become more or less like oxygen. Do you agree with that Steve Garone?

Garone: Kind of. What drives the acquisitions -- and this is not unique to this market -- is that as people begin to adopt a technology or paradigm in terms of development, deployment, maintenance, and management, the comprehensive nature of the solution becomes more and more important.

Vendors who naturally don’t have all the pieces will try to acquire other companies that do -- to will fill in the blanks, so to speak. That’s really what’s going on here and it’s nothing new.

In terms of impact, given the two examples you cited, I think you’re on the money in terms of the ones that were the most impactful, but I think it’s also important to ask the question: impactful to whom? Let me just use the Red Hat/JBoss example to illustrate what I mean. There’s probably very little in the combined solution offered by that acquisition and merger that an end user could not get from another large vendor who provides SOA-related solutions.

It was very important from a competitive standpoint for those two companies to get together, so that they could respond in a way where they could say, “We’ve got a comprehensive answer to the SOA problem that other vendors have.” The impact from that perspective was mostly in terms of the two vendors getting together and being able to compete effectively.

There’s a secondary impact from that particular one, focused on giving more credibility and more visibility to the open-source model, which is playing, as we know, an ever increasing role in the SOA space, as well as other areas. So, yes, they were very impactful, but in that particular case I think it’s important to look at just who it’s impacting and why.

Gardner: I suppose another aspect of this is that those investors who put up money several years ago in the smaller companies that got gobbled up did okay. There was a period of time, not that long ago, when the notion of investing in a software company that was going to be focusing its business on enterprise was a no-win situation. Selling to the enterprise as a startup was difficult. Coming up with a direct sales force was going to be very hard.

SOA, in some ways, reversed that mentality. It has allowed for investment to be made in startups and smaller companies, and has also encouraged companies that may have been infrastructure-focused to adjust their focus, or at least their messaging, toward SOA.

Joe, do you think that the SOA ecology is alive and well in 2006? Do these mergers and acquisitions help that, or do you think that, conversely, when this consolidation phase is over that this investment will dry up?

McKendrick: SOA is something that is not really tangible. SOA is a methodology, a philosophy, and the vendors have a real challenge. Essentially, they’re selling a concept, not a specific product. Maybe ESBs could be example of a tangible product that’s out there, but other than that it’s really difficult to package and market a concept to the market.

With the HP acquisition of Mercury, which had earlier acquired Systinet, I wonder what's going to happen with Systinet going forward. If you look at HP, they’re still compact, but there’s a reminiscence of DEC in there, and Tandem. I could see HP perhaps building Systinet registry solutions into some of its own products, but I fear for this vendor. I think Systinet may just kind of disappear into the large organization, and we may never hear from it again.

Gardner: Systinet had a pretty good run in terms of being a perceived best-of-breed player, even after these acquisitions, so the jury remains out on that.

Let’s go to Jon Collins. Jon, on this consolidation issue, is this a flash-in-the-pan, so to speak, with consolidation? Is it going to just go through its paces and then we'll be back to half-a-dozen large vendors? Or do you think that there is a vibrant ongoing ecological development around those companies that support SOA, either as component services or approaching more of a complete solution?

Collins: Excellent question. I was just reminded, while someone else was talking, a PR person a couple of years ago said to me, “We’re going to have to look at a new name other than SOA, because it’s been around too long. We’re going to have to create a new bandwagon or something, because people are going to get bored with it.” And, here we are two years later. I think we’re only at the beginning of it.

What’s happened this year is that SOA has become mainstream, if one can say that, in the IT industry. Maybe it hasn’t in enterprise companies, but it certainly is in the IT industry. A lot of companies have acquired a smaller or larger companies in order to build their portfolio.

The one thing that’s been lacking so far -- and I’ve talked to a number of vendors about this -- is integration, which is to me the ultimate irony, because its exactly what SOA is about as a concept. It’s about helping bring together the disparate pieces.

One of the things I’ve talked to a few vendors about is the fact that they’re actually having to become service-oriented in terms of how they put together their products, deliver solutions to their own customers, and how they build more specific suites of capability that suits the needs of specific customers. We’re at the beginning of something that is all about how we move away from just having a consolidated set of resources. The point about Red Hat JBoss was well made. Everyone needs all the pieces, and now it’s a case of how they compete in terms of how they deliver the integrated solutions to the market.

Gardner: Okay, let’s go back to Tony. What would be your next pick for a top SOA story or a vendor issue for 2006?

Baer: I’m having a hard time with that, but I think if I were to pick one, it would be governance. That’s a loaded term for obvious reasons, but if you look at the whole notion of SOA, it’s supposed to make exposing your assets, and your data enterprise process assets, easier. If you think about, that kind of goes against the objectives of a lot of the regulations of enterprises.

As we open these assets up, we expose them as services. As we start to ramp up beyond pilot stage, we need control and we need to document what assets are being exposed to services, under what circumstances, to whom, and when. We have to start getting into versioning issues. What’s interesting is that until now SOA has been primarily a skunkworks operation, with few exceptions, maybe for someone like the Verizons of the world.

Mostly we’re focused on runtime governance, which is what services can we expose to which requester at run time, but that was largely in a fairly controlled environment. You’re starting to see more attention to the rest of the lifecycle. As part of that, we’re seeing not just UDDI, Systinet or Systinet cell registries.

We’re also starting to see the emergence of repositories. Something I would not have predicted a year ago is the converging of the two. It would have seemed totally illogical to me before this. At runtime you don’t want a repository with a lot of overhead. Yet I see vendors seeming to building brute-force solutions, just like years ago they brute-forced relational databases, so that their performance was not all that bad compared to legacy databases. So, I guess that in answer to your question, the one thing I would say would be governance.

Gardner: I guess I agree with that. I wondered earlier in the year why this emphasis on governance. It seemed that the value here was that, you’re going to be headed toward some sort of a train wreck if you ramp up SOA, so we have to put governance into play.

Yet as you pointed out, these things remain predominantly as pilots, isolated -- not horizontal, not yet at the scale where the system is breaking down under load, pressure and volume. Yet, there is this big drive to governance. I thought a little bit about that, and I recognized that governance is a way of extending SOA into a larger role of transformation, taking over where policy and even directory left off.

It’s sort of a land grab by these vendors into this larger domain of how to actually manage your business, not just your IT. And it portends the possibility of joining IT governance with corporate governance at some level. So, that’s a big enchilada, a big opportunity. Do you agree with that, Steve? Do you think that this is a land grab, this governance thing, or are they really reacting to what’s a scaling issue?

Garone: I think there’s some of that, although typically what happens in that particular cycle is that a vendor will understand and raise the issue and make it prominent, because they’ve got a solution. Then other vendors will jump in and start creating and talking about nuance that the original vendor can’t deal with -- and what results after that is some level of confusion and lack of understanding on the part of the end user as to how to approach solving that problem. Eventually it all shakes out, of course, but that’s usually the cycle that’s takes place.

If I were going to talk about some of the major highlights of SOA in 2006, that brings me to one point that I think stands out. There’s been a lot of talk about enablers to SOA -- governance, ESBs, which I think is another very important one. Because of the way vendors approach providing solutions, sometimes the enablers become a point of confusion to the point where it actually slows down adoption.

I think some of that is taking place with SOA. I think some confusion around standards is also playing a role in that. I forget who it was earlier who said that they feel we’re right at the beginning of SOA adoption, I agree with that, and when we talk in a little while about trends in 2007, I think that’s going to come out more clearly.

So, yes, I think that vendors are trying to do a land grab, but in the process they may be introducing a level of confusion that may actually be slowing down adoption in a big way.

Gardner: How About you, Jon? Do you think that this movement toward governance and the alignment, if not mashing up, of repository and register, is this just a reaction to the potential scale issues of SOA or is this really trying to broaden the purview of SOA into a higher business set of issues?

Collins: I think it’s broadening its horizons. SOA, certainly from my experience, came out of the developer world, in terms of evolving out of component-based development and out of the Web-service based application space. It’s really a post-deployment characteristic of enterprise architecture, and if that’s the case, then it needs to built in. There’s operational management. There’s deployment management. How do we cope with this stuff once we’ve got it? The idea of a registry, the idea of a repository of assets, all of these things start to kick in.

It’s broadened out from it’s original space and it’s gone into a place where governance is much more necessary, but equally it’s a symptom of maturity. A lot of IT starts in terms of innovation, where people are pretty much left to work out how. Networking started with people saying, “Hey look, we can join this to this,” and then suddenly everyone’s joining everything to everything. After that, people are saying, “Shouldn’t we start to work out whether or not we should be joining things together?” And there’s a level of maturity that I think we’re hitting with SOA, which means that governance becomes far more important.

Gardner: I suppose that if eventually you have to get the sign-off from a CEO or a very high-level executive on investing in SOA activities or continuing to invest in integration at any level, it might be nice to walk into that leaders office and say, “Hey, not only are you going to get some more efficiency and some agility and faster time-to-market for your applications, but we’re going to give you a dashboard where you can actually go in and see what’s going on, vis-à-vis IT into your business processes. We might even give you a couple of dials and knobs where you can start turning up and down or prioritizing.”

Over to you Joe, can we use this governance thing as a way of bringing to the chief executives of these major corporations more controlled visibility into IT -- and therefore into their businesses?

McKendrick: Absolutely, Dana. In fact, look at most enterprises. They’re not going to take any projects seriously unless there’s some aspect of governance. By governance I think we’re talking about two levels here. There’s the nitty-gritty deployment, management, registry, repository, versioning aspect of what’s being done with the artifacts, and with the services that are being put out there. Then there’s the governance we speak of, which involves bringing in the business and asking the business what they need out of IT, what they need out of the technology.

Be it a committee or an evangelist within the enterprise, you need some type of board, body, committee, or team that represents a cross-section of the business that can look at what they need and interact with the IT department to bring about the solutions that are required for the business. IT can’t operate in a vacuum and doesn’t. It’s not IT’s fault, but they can’t have the sense of what a business needs, or what the business units need from week to week or month to month. You need that business input and that the high-level definition of governance. That’s the only way SOA can be sold into the organization.

Collins: I totally agree, and furthermore this is a fundamental re-alignment. We’re actually seeing the business itself take more and more of an active role in these issues, and talking about things like IT risk with relation to the overall business risk landscapes. IT is actually seen as a valid part of the businesses, rather than seeing IT -- as often it’s seen – as a little place with a bunch of geeks developing stuff. So, yes, absolutely, and it’s being driven by both sides.

Gardner: All right, let's move on to our next big issue from 2006. To you, Steve Garone. You mentioned ESBs, and you mentioned the possible negative impact of over-emphasizing -- not the main course, but some of the side dishes for SOA. So my question to you is, is ESB your next big story of the year? If so, do you think that we’ve watered this one down, that people don't have a firm idea of what ESB’s are? What’s the story around ESB’s this year?

Garone: It’s fair to say that ESBs were at least one of the top stories in 2006 in relation to SOA. That makes sense, because as people started to adopt a SOA approach, whatever that meant to them, they were searching around for the right infrastructure elements that they needed, in order to make that architecture work for them. The notion of an ESB is right at the core of that. In fact, when I was thinking about what we were going to talk about today, the two issues on my list that stood out the most were governance and ESBs.

Some of the confusion around ESBs has stemmed from the fact that they’re pretty big items in terms of what they’re capable of doing. The reality is that a lot of organizations may not need a full ESB, as many vendors are defining it, to make their environments work. They may already have some of the integration capabilities. They may have some of the event-handling capabilities in place, and they really may just need some mediation and governance elements added in, in order to make their environments work. That’s introduced a bit of confusion that has caused some grief amongst end users, at least the ones that I talked to.

There’s a tendency also to look in the other direction in terms of sort of creating an “Uber ESB,” if you will, one instantiation of that being in the form of a JBI implementation that will allow people, regardless of how they’ve chosen to implement the functionality, to bring that all together and integrate that into a complete environment. What you then get into is, “How do I navigate the world of standards to make sure that I’m adopting this and implementing it in the right way.”

So, ESBs have been very prominent. There’s been a lot of noise, news, and discussion about ESBs, a lot of product announcements in 2006. I think you’re going to see that continue in 2007, hopefully in a way that clarifies and brings together the activities to something that’s coherent and therefore makes organizations feel comfortable that they can move forward.

Gardner: We saw, just recently, announcements -- roll outs, I suppose, more appropriately -- from IBM and JBoss around ESB. There’s been some open-source activity. We’ve got a couple of Apache Foundation projects that are ESB-related that IONA’s active with. The question to the rest of the group is, do you see ESBs remaining standalone? Does this become something you take as a component, as a best of breed, or is this something that should be baked into a larger solution, approach or a technological approach to SOA?

Baer: I would vote for the fact that it becomes part of your general platform, your general infrastructure. I don’t see ESB as being an end. It’s a means to an end, and as Steve was saying, there are many different functions that an ESB can or should perform, and that’s going to depend on what your existing IT infrastructure is. It’s going just like governance. It’s going to become an extension of whatever your service platform is.

What’s very interesting is that in the past year IBM finally admitted that ESBs are not just an architectural pattern, it’s actually the real product. There’s been a lot of confusion, fear, uncertainty, and doubt out there as to what ESBs actually are. What I would love to see -- and of course this would probably never happen -- is some industry consensus, “Here’s a Lego blocks model. Here are different things that an ESB can do.”

But, guess what? In some cases, if you already have TIBCO in your environment, or something like that, you may not need, for example, the capabilities that a lot of ESBs have. So, I would love a building-block approach.

Gardner: We can probably loosely define ESBs as EAI, but with more hooks and on-ramps and off-ramps.

Garone: Hooks and standards.

Gardner: Hooks and standards, there you go.

McKendrick: To Microsoft the ESB probably meant enterprise service buses or those vehicles that employees ride around on at the Redmond campus. It seems that Microsoft has gotten on board with the ESB concept lately as well, which surprised a lot of observers, because Microsoft usually likes to take its own route with things and not buy into the acronym of the moment.

Gardner: That would be more hooks than standards then, wouldn’t it?

McKendrick: Probably, yes, highly integrated within .NET Framework and the Windows Communication Foundation.

Gardner: Moving on to standards, Joe, you had an interesting blog the other day about the WS.* standards. And I wonder if you thought that that was perhaps an issue by omission. It seems to me that there was a certain atrophy in 2006 around SOA standards. We did have BPEL come out with a 2.0 major release, upgrade, improvement -- but what’s the status of specifications and standards for SOA? Was this sort of a lull in the action this year?

McKendrick: Well, Dana, most of the standards focus on web services, and there are folks out there that would say SOA isn’t just about web services. It’s probably 90 percent of what SOA is about, but most of the activity is been around web services. Some of the standards are highly adopted everywhere you look, of course XML being the prime example of that.

You have SOAP and then WSDL, which are ubiquitous everywhere you go. The WS-splat or WS-* standards have been another story. There’s been a lot of controversy, a lot of confusion. In surveys I’ve done, I’ve seen some very low levels of knowledge -- let alone adoption -- of the WS-splat array, with the exception of WS-Security, which has been taking off fairly nicely and proving itself in many situations. You’re right Dana, it has been a build year/build season for the standards bodies this year.

Gardner: Jon, do you concur with that, that the standards had evolved and it was time for the implementations to catch up? And, should we read anything more into that?

Collins: That’s probably a good way of putting it. The other standard I’d bring into it is the management side of things. While there may not have been many great advances in the standards themselves, it’s more of an aligning of vendors behind the various standards or an agreement that there’s going to be a interoperability between them.

If you’ve got, let’s say WS-Security or WS-Policy, because of the nature of SOA and how it actually needs everything to be ready before any parts of it can work, that’s a very broad statement. But if you’re trying to manageably deliver multi-vendor, multi-application, multi-system environments in an enterprise context, a lot of these things are there already.

While some standards that were actually quite advanced to be able to roll out full fact SOA, a lot of other standards are still in catch-up. We’ve got standards that are saying, “Come on guys, let’s get out there and do the job.” Meanwhile, things like management are saying, “Hey, we’re coming. We’re on our way. We’re a bit behind. We’re still try to work out of what a CMDB is and how to talk to each other, but we’re on our way and will be there shortly.”

Gardner: All right, before we move into the prediction’s phase, I wonder if anyone else out there had a major issue from 2006 they wanted to raise as one of significant importance.

Collins: I’ll just say one word about Microsoft. Maybe the fact they even mentioned SOA was a pretty good incidence, if not a major happening.

Baer: Actually, I’ll add one thing to that which is that Microsoft has even felt compelled to publish a web page listing which standards they will support. When would you ever have thought that Microsoft would’ve ever published such a page in the past.

Gardner: Probably the same time they came out and supported Linux.

Collins: It was within a month, wasn’t it?

Gardner: Well, as a matter of fact Jon, we did a recent BriefingsDirect SOA Insights Edition on the SOA-Microsoft question. So, moving into 2007, next year, we’ve talked about some levels of maturation. We’ve talked about some technology developments, some vendor activity, but the question still remains is when do we move SOA -- and I do emphasis "when" and not "if" -- from the pilot and skunkworks-type of approach to, “This is how we’re going to do more of our IT, more of our development, more of our process in modeling. We are going to build this out?”

So let’s go around. Is 2007 the year for the SOA build-out, or are we being a little bit too optimistic? How about you, Steve?

Garone: It will be the beginnings of what you’re terming as a SOA build-out, but I’d like to deal with less of a product- or a vendor-related issue now and talk a little bit more about end-user organizations themselves. I’m still amazed, somewhat amazed at how much I’m hearing organizations talk about two issues.

One is that we hear about IT, business analysts, and business people working together to come up with requirements and actually having tools where they can work together to build business processes and then implement them and support them with services. But, little of that is actually going on and being effective within organizations. I hear a lot of folks on both sides of that pointing fingers and claming that the other one doesn’t get it, and that’s sort of manifesting itself in several ways.

It’s partially responsible for the pilot-nature of implementations that you see today in most of our organization that have tried SOA. I think it’s also responsible for the overall slowness with which you see this moving through some organizations. Over 2007, I think you’re going to see some shake-out and some advancement in terms of building that bridge between the two organizations within corporations that want to do SOA and therefore forming the foundation for moving on, but I don’t quite think we’re there yet.

Gardner: So, for 2007, if we could put some sort of a metric on adoption, with 1 being no further movement from 2006, and 10 being a total SOA in place, architecturally across the board, where do you think the dial, the meter, the needle, will go to in 2007?

Garone: There are two elements to that. Organizations that have not done it yet are going to, for the most part, remain down at the 3 or 4 level, if 3 or 4 level is defined as doing small-pilot implementations. Some percentage, say 25 percent to 30 percent of those organizations that have gotten to that first stage, will move on in a big way. I still believe that there are some impediments, having in some part to do with organizational issues and the relationships between IT and business, that are going hold that up a bit.

Gardner: Just to poke a little fun at our own industry, the IT research business, we can safely say from reports that have come out in 2006 that somewhere between 10% and 90% of enterprises will be doing SOA.

Garone: Well, yes, that’s true.

Gardner: What are your other predictions for 2007?

Garone: Getting back to one of the issues that we talked about earlier, the whole ESB issue and its place in an overall infrastructure that supports SOA will get pretty much resolved and be much clearer to end users as we get toward the end of 2007. I also think that there’s going to be a lot of work around registry, repository, and standards, mostly because as organizations start really using SOA to support real business processes with services, there’s going to have to be a lot in the way of semantic-level information and defining relationships that are going to be have to be incorporated in a real robust repository. So, I think from the product standpoint, those are two areas where we’re going to see a lot of advancement.

Gardner: Okay, how about you Tony? What are your predictions for SOA in 2007?

Baer: A couple of things. First, a more general trend, which is not SOA specific. We're no longer in the era of big-bang implementations. So, in terms of the pace of SOA adoption, it’s going to be a pretty boring answer: It’s just going to continue to be incremental.

You’ll see several hot-spots in organizations that are dealing with certain consequences of M&A, for instance, where they need to link to general ledger systems or something like that. You’ll see some opportunistic SOA there. Otherwise, it will be that the leaders will continue, those that have been leaders in 2006. Like the Verizons of the world. I hate to keep harping on that. I saw them working outside my house yesterday. So, I figure that they have some ambitious initiatives on the table there. But, I think the mainstream will incrementally get to the first things beyond the skunkworks stage.

Gardner: How about on a scale of 1 to 10, with 1 being no further movement from 2006 and 10 being full metaphysical SOA implementations?

Baer: I was going to say about a 3 or 4, just averaging things out. The Verizons of the world may be at about a 5. Whereas say the mainstream was probably at a 2 or a 3, but that still is progress.

Gardner: Okay, Jon Collins, what are your predictions for 2007?

Collins: Predictions for 2007: The thing that we’re going to see the most of is what we’ve talked about already as industry acceptance of SOA in 2006. In 2007, we’re going to see an enterprise acceptance of SOA. In terms of adoption levels, people aren’t going to suddenly sort of rip out everything that they’ve got and say, “Let’s paint the whole town SOA.” But there will be more of an understanding of where it can benefit the organization.

When there are projects that are appropriate, they will be done in a way that encourages and helps develop more service orientation inside the organization. That fits in with the need for and the delivery of improved governance of better relationships between IT and business, a better understanding of business issues on the technology side. So, it’s everything, all moving forward at once, and without that you can’t really do SOA.

The other thing is that it does go all the way down the stacks. We could talk to SAP. I’m talking to clients at the moment who are doing SAP rationalization projects. We could talk to HP about how they’re providing a platform to support applications, and talk to Cisco about Service Oriented Network Architecture (SONA). If you’ve got something that can be understood as a service delivery, then they’re putting in place the network facilities to support that. Then, you can even talk to the service providers, the British Telecoms, who are putting in place some service delivery networks.

It goes all the way down the stack, and I think we’re going to see that integration, top to bottom, become more of a reality in all those companies that I’ve just talked about there. They’re doing it for themselves right now, and they’re going to start doing it with each other, which will be a change.

Gardner: So the good news is that SOA affects everything and the bad news is that SOA affects everything.

Collins: Absolutely.

Gardner: On your scale of 1 to 10 next year?

Collins: I put it down as the difference between acceptance and implementations. I’ll still put it down as between 2 and 3. It’s low, but it’s a good low.

Gardner: It's 20 percent, it’s double digits.

Gardner: Joe McKendrick, 2007?

McKendrick: SOA is a very amorphous concept. It’s not a tangible concept, and SOA as a term has crested, as we’ve discussed in this conversation. It wouldn’t surprise me if we started seeing more vendors move toward some other terminology or other focus in their marketing efforts at the vendor level. Maybe we’re going to be hearing more about EDA, Event Driven Architecture. It has more of an action sound to it. It’s something new, and it has that real-time aspect that ties into business intelligence, and can be integrated well within business intelligence solutions.

So, I wouldn’t be surprised if we saw vendors moving away from SOA terminology and toward things such as EDA, maybe software as a service. Maybe we’ll be hearing more about that. At the enterprise level, the end users in the trenches, where the real work takes place, there’s going to be slow, steady progress toward the SOA model. It’s not going to be a revolutionary thing happening in 2007. It’s going to be evolutionary. There are going to be more pilot projects moving into the enterprise production space. What may mitigate this, and what would be interesting to watch over the coming year, is the availability of staff, availability of the right skills.

The economy has been going well. I don’t know what 2007 will bring, but by all indications, IT budgets are holding up and there’s been already shortages of the right types of skills. For enterprises that need to keep their day-to-day operations going, it maybe difficult to allocate the human resources to really move SOA projects forward.

Gardner: On a scale of 1 to 10, increase in SOA adoption in 2007?

McKendrick: Two, a solid 2.

Gardner: We’re going down, we’re going down. I’ve got a few predictions myself that I'd like to throw out. I think we can also look to 2007 as being the year of modeling. I think we’re going to see a lot more product and specification and methodology around the modeling of services that can leverage and exploit some of the activity from 2006 around this governance and the registry and repository. I’m expecting some much improved tools, graphical tools, and some innovative approaches to how to choreograph -- and that plays into your event-driven stuff, too.

I’m also predicting that this notion that Web 3.0, which is better characterized as the "Semantic Web," will offer on one hand competition against the SOA messaging and value, but more importantly offer a tag-team approach. We’ve discussed this a little bit before around Web 2.0.

If we do start getting some specifications, and as IP Version 6 becomes more widely deployed around the world, we’ll get much richer in approaching this level of a semantic Web capability on the network. If you can combine that with what you’re doing internally, that that’s one of the big tipping points for the further adoption of SOA.

I’ll use that as my opportunity to say I see about a 4 in terms of actual SOA adoption increase, so 40 percent, which is pretty big. I’m going to go on the top side of this trend for our group.

I’m also going to go out on a limb and say that I’m going to expect some big SOA-related acquisitions again in 2007, and I’m going to look to SAP and Oracle as likely candidates for acquiring other companies. The net-net on that is the application guys are going to start to flex their muscles on SOA. If you have both the infrastructure and the applications, and then you produce the SOA benefits, that’s a very advantageous place to be.

That’s it from me. Anyone else want to chime in on 2007?

Garone: Yeah, I want to respond to a couple of things. Let me start with what you just said around the enterprise apps vendors. That was one 2007 prediction that I had on my list, driven by something a lit bit different than what you said, though. One of the empirical pieces of feedback that I get often from end users is that they’re waiting and pining for the day when they’re not wrapping their legacy apps anymore in Web services standards and, all this stuff will be essentially done for them by the people who provide the applications.

Gardner: SOA-native-as-services.

Garone: Native as services, exactly right. I think you’re going to start seeing the SAP’s and the Oracle’s and everyone else that services that market, stepping up in a big way to help solve that problem, so I completely concur with you on that.

One of the proof points that I didn’t mention earlier -- I’d mentioned it in a previous podcast -- is the lack of synergy, integration, and operability between the business side and the IT side in real life. Most of the end users I talk to, when I see them present, when I talk to them in detail about their SOA activities, they really just cite cost savings. They mention raw cost savings in terms of development and operations as really the primary, and in some cases only, benefit of doing an SOA implementation.

We all know, as analysts and people who have talked a lot about SOA and how they can benefit organizations, that that’s a very narrow view. What that tells me is that the business case, the true comprehensive business case, still hasn’t been met in a lot of cases, and hasn’t been articulated very clearly. I think that’s going to start turning around in 2007.

Gardner: When you emphasize the modeling and the applications, you start getting closer to that business-level benefit?

Garone: I agree. The other thing I would just say very quickly is that -- and I’ve mentioned this before – it’s going to be harder and harder to divorce the services approach to building solutions from the virtualization area that a lot of companies are looking at, because they’re looking to do them at around the same time, and they’re very closely related.

So, we’re going to see more of a comprehensive view of that, not only in terms of how organizations want to implement it, but from vendors as a response. I also think you’ve defined a good podcast topic for the future in Web 3.0.

Gardner: We’ll be there. All right, gentlemen, thank you very much. Let’s quickly go around the table for purposes of disclosure. If there are any vendors that were mentioned or implicated today that you do business with, it would be a nice time for you to mention that, so that we all have a sense of where we’re coming from. Why don’t you go first, Steve?

Garone: At this point I would just list IBM.

McKendrick: IBM, and I’ve also done work with Systinet.

Collins: Mercury, so HP I guess.

Baer: Mercury, ditto.

Gardner: I’m going to disclose that I’ve worked with Borland, Cape Clear, Eclipse Foundation, Hewlett-Packard, IONA, and I think that’s probably it.

All right. Thanks very much. Also, if any of our listeners are interested in learning more about BriefingsDirect B2B informational podcasts or to become a sponsor of this or other B2B podcasts, please fill free to contact me, Dana Gardner at 603-528-2435. That’s it for this week’s, and I suppose this year’s, SOA Insights Edition. I’m your moderator and host, Dana Gardner. Thanks for joining.

Listen to the podcast here.

Transcript of Dana Gardner’s BriefingsDirect SOA Insights Edition, Vol. 6. Copyright Interarbor Solutions, LLC, 2005-2006. All rights reserved.

Wednesday, December 13, 2006

Transcript of BriefingsDirect Podcast on IT Consolidation Trends and Methods

Edited transcript of BriefingsDirect[TM] podcast on IT consolidation with host Dana Gardner, recorded Oct. 31, 2006. Podcast sponsor: Hewlett-Packard.

Listen to the podcast here.

Dana Gardner: Hi, this is Dana Gardner, principal analyst at Interarbor Solutions, and you’re listening to a sponsored BriefingsDirect podcast. Today, a discussion about IT Consolidation, taking a distributed, geographically dispersed set of datacenters or application infrastructures and bringing them into a more rational approach -- to get more for less, in less places, with less labor and less total cost. The goal is to eventually take this consolidation effort to a strategic level and move to a full-service infrastructure and allocate IT resources efficiently and with more flexibility.

We are joined today by two executives from Hewlett-Packard’s Consulting and Integration Group: Ewald Comhaire, director of the global practice for Infrastructure Services, as well as Peg Ofstead, worldwide solution lead for HP IT Shared Services and IT Consolidation Services. Welcome to the show, Ewald and Peg.

Ewald Comhaire: Thank you.

Gardner: Let’s address the rationales for IT Consolidation. Does this vary from site to site, from enterprise to enterprise? Is this a highly individualized approach? How do we implement IT Consolidation? Are there some uber trends driving this? Let’s start with you, Peg.

Peg Ofstead: The desire to consolidate differs from company to company. Our best practice is to first clearly discuss the objectives of the organization. Cost has been a driver for many years. There are other issues that drive people to consolidation. For example, some people run out of space in their datacenters; others have issues with power or cooling that cause them to replace their technologies.

For others, it’s a case of simplifying; they just have too many data centers, too many different kinds of technologies, too many redundant applications. They want to simplify their environment. Ewald, do you have some other thoughts?

Comhaire: Yes, there is definitely the cost element, and doing more with less is important to every customer. We have been doing this for 10 years. We see customers with more balance between these cost elements as well as improving the quality of the services they deliver and how fast can they turn around the request from the business.

Reducing risk by reducing issues with security, or improving business continuity and availability of the solution is a large objective we see across the industry. Each customer is unique, but we see these drivers most of the time.

Gardner: Obviously economics and cost issues are big drivers for this, but it wouldn’t be possible without a certain level of technology innovation and capability. You need to be able to consolidate by managing the complexity of the hardware and the networks. Can you give us some history of IT Consolidation and explain what it is technically that is allowing this to be accomplished now?

Comhaire: I’ll be happy to start. When we look back 10 years ago, most of it was driven by the capabilities of the infrastructure. As you consolidate you need better processes, you need to manage consolidated infrastructure, and that adds complexity. Do your tools, your systems, your software, your operating system have the capability to really consolidate on a technological level?

Fewer customers are worried about whether the technology can do it. Other factors are now driving the need for consolidation. Customers are looking at a functional level, consolidating not around the technology, but around functions being offered to the business, using that as a central element, with the technology following the service.

We have seen an evolution of what customers take as a primary centerpiece for consolidation. It has moved off technology and more to a functional type of consolidation. Peg, anything you would like to add?

Ofstead: Also, people are consolidating datacenters because telecommunications is so much cheaper and better now. In the past, they couldn’t because of latency and other issues. On the technology side, especially in the Microsoft Windows environment, the virtualization technology now allows people to be more aggressive in that space.

Gardner: The network is now able to support this consolidated effort. This also has to do with the shift that the Web has brought in terms of centralized applications and browser-based interfaces. Over the last 10 years there has been a significant shift to Web-type applications. Portal applications with browser interfaces have also been an accelerant to this move to IT Consolidation. Do you agree with that?

Comhaire: Absolutely. Where it has really helped is with datacenter consolidation. Because of the lightweight graphical user interface (GUI) that these Web-based applications provide, you no longer need to have all the servers -- or at least part of your application architecture -- located close to the end users.

With the latencies and improvement in networking, you can now do what HP, for example, is doing with its own datacenters: co-locate them in one location of the world and still have good response time for all their users. It would be much more difficult if you have a complex client/server architecture, and the need to move a large amount of data to different parts of the application. It would be very tough to do, and this definitely helps.

We also have "lights-out" management where management capabilities are done remotely, 24x7, in different parts of the world. There are countries in Eastern Europe, the Middle East, China, and India where you have some capable resources that can do remote management, but you would not typically put your datacenters in those physical locations. Datacenter consolidation enables you to utilize the capabilities in remote management areas when your datacenter could be in a place like the HP datacenter in Houston, Texas. Consolidation has helped with this management significantly.

Gardner: Your expertise, your people, can be in one location -- but the physical plant can be in another. You can look for a lower cost of real estate, energy, and safety, when the plant is away from high-risk areas. These all play into the IT decision making about IT consolidation?

Comhaire: Absolutely.

Ofstead: That is exactly what we’re seeing.

Gardner: Now that we have discussed how the Web has changed the nature of applications, and the technology from the network perspective has given us the freedom to decide location -- do we see anything on the horizon that will change this even more? Are there other factors that make the decision-process around consolidation change again? Are there convergence or virtualization issues?

Ofstead: We can assume technology will continue to improve. Look at the cost of storage today, for example. People forget it’s not just about technology.

One of the key factors to keep in mind when you’re consolidating is the human factor. We discussed being very clear in your objectives. It is important that your stake-holders buy-in on your objectives, and that includes your line-of-business partners. There could be issues like the line-of-business may own the datacenters, or may own the servers that you’re trying to consolidate. Their buy-in is crucial to your ability to successfully consolidate.

There is also a buy-in of your IT staff, many of whom can be impacted during consolidation. If you are eliminating legacy technologies, it may mean they need to be re-skilled, or if you are closing datacenters, it may mean a larger impact to their jobs. It is very important to consider the human factors and the governance issues related to consolidation that go beyond technology.

Gardner: Obviously this is not a trivial, fly-by-the-seat-of-your-pants decision process. We have issues around ownership and a culture of control and the cost of real estate in terms of applications, and around how to rationalize data itself within a data structure. How do companies begin to approach this "hairball" of interrelated dependencies and complexities once they have begun to take IT Consolidation seriously?

Ofstead: Ten years ago, we were seeing the consolidation of UNIX servers. The largest companies we worked with were telcos, and they often had hundreds or thousands of UNIX servers and were consolidating those. Companies in all industries have been basically doing server consolidation and storage consolidation for many years now.

In the past few years, they have tackled the harder things. Rationalizing the application and information portfolios is much harder work. We saw very little of that five years ago, but it is more prevalent today. We are seeing more datacenter consolidation and relocation of datacenters as people consider issues like Sarbanes-Oxley, which especially impacts the financial institutions.

Gardner: We have also seen waves of consolidation at higher business levels between mergers and acquisitions, particularly in telecom and a number of other industries. I think that would also bring into play massive infrastructure redundancy that needs to be addressed.

Ofstead: Absolutely. Companies that have already gone through a consolidation acquire a firm and suddenly they have twice as many datacenters as they need. They have an overlapping application portfolio, so this is an ongoing process for many firms.

Gardner: Ewald, can you give us a rundown of the methodological decision process? Large enterprises, regardless of their situation, are facing this complex proposition. What needs to be considered? How do you approach this from a professional services perspective?

Comhaire: The first thing we do is start with the strategy and analysis phase. In this phase, we conduct a workshop to clearly define the business goals with the customer. We bring in our 10 years of consolidation experience. We help customers to think broadly, not just in terms of servers, but about the number of workplaces they have, whether we can do workplace or remote office consolidations, their networks, and their higher-end services like data warehouses.

We provide a pretty broad view. We help to pin down the customer on key matrices around the success of the project. What is in scope, what is not in scope, and how this IT consolidation aligns with the business. We work with the IT parts as well as making a link with the business, and use it as a starting point for the project.

We then do a rapid opportunity analysis for the domains that are good candidates. We are looking for a good return on investment (ROI) for the customer on a high level without doing a detailed analysis.

We then do an inventory and discovery. Many customers have an idea of what they have, but not enough to do a real ROI or total cost of ownership (TCO) calculation. We provide some agentless discovery tools that show the workloads as well as the assets there -- how well the equipment is utilized, what workloads are compatible or can be consolidated with other workloads, and how many databases are there -- so we understand by function what is used today and what are some of the possible economies of scale. This is typically done in the first phase.

Gardner: It sounds like you need to step back and look at what you actually have, and then what you want to have, functionally, going forward, and then to try to reconcile the two.

Comhaire: Yes, that is important because these projects typically take quite some time, and a consolidation project is never done. We are always looking at the next thing we can improve on. At some point you need to have closure on a certain part of the project.

Phase two is the architect and validate phase. Then you can think about a high-level architecture for the desired date. We now know the objective of the customer, we know the “as is” state through the discovery, and we can now do a high-level design for the desired state.

At the same time we are validating the ROI model. If there are cost-reduction objectives, and the customer wants to have those in hard numbers, we can do a full ROI analysis and be confident that we can get the ROI that the customer was expecting. That’s pretty much our second phase in the consolidation methodology.

Gardner: I have to assume that we’re starting to tinker with some mission-critical applications and there is some concern about the continuity and risk of disruption. On one hand the company would ask, "Are we going to have downtime? Are we going to have interruptions? Will this be seamless to our end users, both inside our organization and our customers outside?" How do you address the issue of confidence? How do you guarantee this is not going to be a disruptive activity?

Comhaire: Great question. Our third phase is the detail design and planning phase. Here we do the real planning of the implementation and all these things are addressed in the detailed level design. For example, if we have a datacenter relocation, what methodology are we using to move to data consolidation?

These design criteria have an impact on the downtime for the business, and this is where all the expertise comes in around migration of data and applications. We also make sure that the target environment -- as it’s being designed to handle multiple coherence and co-located workloads -- is bringing the availability up and not down, which is an inherent risk of consolidation.

We have to write a business continuity design so that the desired state has much more tolerance against disasters than the existing state. We know in phase number three that there is a positive business case. We are then allowed to go to a level of detail that is very much implementation-oriented.

In the second phase, we typically want to do the minimal design to guarantee that the project will have a good ROI, so we don’t go down to this level of detail. But in the third phase we do to make sure we get the trust and the minimum interruption for the business through the right choice of design methodology.

Gardner: Peg, on this notion of risk reduction, do you feel confident enough in doing this many times that you can almost, and perhaps actually, guarantee continuity? Do you go in and say, “As part of our contract we have our own agreement of certain levels of service,” and do you go to the bank with that and say, “We’re going to make sure your businesses continue?”

Ofstead: Absolutely. That is part of what we do with consolidations because we are working with mission-critical workloads. In every region, we’ve worked with clients that have been in those situations.

Comhaire: We also have a very strong mission-critical support that’s been highly rated by Gartner year after year. That is really a true differentiator for us in those large consolidation projects because we can give the customer the assurance that we can manage their consolidated workloads at the levels of availability required.

Gardner: This is an issue where failure is not an option.

Comhaire: Exactly.

Gardner: Also on this notion of business continuity, companies are dealing with a separate area of risk: Where to locate for disaster scenarios -- whether they’re weather-related, disruptions from energy problems, or perhaps even war and terrorism issues. It seems that companies need to take that into consideration as well now.

When they think about IT Consolidation, are they also thinking about the ability to maintain business continuity, regardless of external threats?

Ofstead: Absolutely.

Comhaire: Absolutely. Yes, Peg.

Ofstead: When we do datacenter consolidation, even if the company had a wonderful business continuity plan and could execute it well, you essentially break that when you change the datacenter. You have to start over and state the business continuity plan. That is very important. We’re seeing some institutions, especially financial ones, moving the datacenters to safer locations because of Sarbanes-Oxley implications.

Comhaire: It sometimes does trigger some finer discussions with the customer, because most would expect that with a consolidation, including cost reduction, that they can also get the whole business continuity and disaster recoverability free of charge as part of the consolidation project.

Sometimes we have to stress that this is really an added functionality. While the pure consolidation project will return a significant amount of savings, there may have to be some investments kicked in as well to address the business continuity element.

Gardner: Perhaps when you embark on this journey -- to kill two birds with one stone -- you look at your ROI/TCO issues from a consolidation perspective, but also look very carefully to your business continuity issues in terms of safety, reliability, and disaster-prevention or -recoveries?

Comhaire: Yes, if you look at eight out of 10 of our success stories, and you look at the advantages that the customers cited after HP helped them implement the consolidation project, almost everywhere you will find better performance, better availability, and better disaster recovery -- meaning that in eight out of 10 projects it is addressed effectively with the customers.

Gardner: We’ve mentioned regulatory issues a few times, particularly Sarbanes-Oxley in the United States. Help me understand the implications here. How does a regulatory imperative affect the decision making, and even perhaps spur on these IT consolidation activities?

Ofstead: Sarbanes-Oxley has some regulations that specify the availability requirements for financial institutions, and how quickly they must recover certain transactions. That happens when they are audited. That is why the financial institutions are looking for more secure datacenters and more robust disaster-recovery capabilities. Europe is in a similar situation with the Basel II regulations.

Gardner: Anything to add to that, Ewald?

Comhaire: We have software that helps customers meet regulations like compliance with standards, specifically on the IT side. We have tools in the HP OpenView portfolio with respect to software distribution. We can control exactly what needs to be running on servers and desktops.

There are some interesting offerings that we have on that phase, but they are second to the main objectives of a consolidation project. We have not seen many projects that are driven by regulations. It is typically an element of a consolidation project, rather than the main driver for it.

Gardner: We have looked at the business case and rationale for doing this, and at the history of why the technologies and network resources allow this to happen. We’ve looked at some of the tangential implications around people, process, and business recovery regulation.

Let’s talk about the payoffs. We have seen some publicized information around HP’s own multi-year approach to consolidation and significant reduction of total costs. Do you have any metrics or case studies or anecdotes that will give people a sense of the magnitude of the shift to benefits here?

Ofstead: Obviously the payoff is going to depend on how diverse your IT environment is to start with. If you had multiple acquisitions, as HP has, there are more payoffs because there is more to clean up. Large companies that have not yet consolidated their servers can probably get a 25 percent to 40 percent TCO savings around those server farms and their storage.

You get real savings in the applications phase. Far beyond TCO savings, you get business benefits. For example, HP is going to a single data warehouse, consolidating over 700 data marts. The business benefit of having a single source of data and consistent information about our customers and our orders is that we have huge ramifications that go way beyond just IT savings for a corporation.

Comhaire: Also, the cost model that we use is a simple one. It can be well understood by the customers as well as HP.

We use four primary dimensions. One is cost and what we can do on cost saving and cost avoidance. Avoidance is an increasingly important element. The second is the quality of service. Did we improve the quality of the services that were delivered to the business? That has some business value as well. The third, are we able to do things faster, turn around the request of the business in a faster way, deploy things faster, put changes into place faster.

The final one is have we reduced, in general, the risk position of the customer, whether it’s against security, or intrusion at the level of compliancy, or at the level of continuous operation and disaster recoverability. We’re putting value on whether we reduced the risk with the customer.

There are some great anecdotes. We have had some large-scale consolidations where, obviously, we go application-by-application and then find that these applications need to be recompiled, that the customer has lost the source code. This was a risk that the customer did not realize before they did the consolidation project.

Now we realize there are some applications where we are at risk because we no longer have the source code, and we were able to reverse and re-engineer that, and get back to a solid code base so they can move forward. We now better understand the risk, we measured it, and we are also able to handle some of the implications of that so that is no longer an issue. It’s an anecdote, but it happens more than you would think.

A customer has said they had 500 servers when they did the whole analysis. It turns out they had close to 1000, and they were surprised there was such a big difference between what they believed they had and what they really had. So the whole ROI model was actually changed overnight based on that discovery.

Gardner: I suppose that this process does unearth these various cans of worms, but in doing so, you’ve got a much more coordinated, managed, professional inventoried approach to your code and your assets. You can move forward with a strategic approach to IT rather than a piecemeal or ad hoc approach.

Comhaire: That is often underestimated as a benefit. It is difficult to say in a customer discussion that as part of the consolidation exercise we will clean up some of these legacy things that you may not realize have built up over the years. It is difficult to say up front, but often when customers look back, that is a major achievement in the project.

The nice benefit of it is they now have a good handle on their assets, the workload they have, the issues that were uncovered and are now clearly on the table and have been resolved. They have a much better foundation to build on for the future.

Gardner: Perhaps this is a difficult and cathartic process, but with recurring dividends over time. I suppose that IT Consolidation will now impact such trends as virtualization and service oriented architecture (SOA), where it will make tremendous economic sense to have a full-service infrastructure approach.

If you’re going to consolidate, you’re going to want this infrastructure to support legacy, distributed Web, and a variety of different types of application sets, depending on your vertical industry. Is it too much of a pie-in-the-sky notion that we can have a single -- or at least a significantly reduced -- number of infrastructure approaches to support a variety of business processes, business activities, and applications?

These days with the impact of virtualization, with increased use of open standards, with increased use of open source, are we finally at a point where we won’t have monolithic approaches to a variety of applications, but perhaps more of a horizontal, rationalized infrastructure to support more and more types of business activities?

Ofstead: Certainly consolidation is the opportunity for folks to move to a real strategic platform view. While they’re cutting costs, they have the opportunity to really move to new technologies and virtualization. For example, we’re seeing a lot of people using Linux. One of the great things about that is not only are you able to clean up your environment, but you are actually able to move it forward to new strategies. Ewald?

Comhaire: Absolutely, Peg. The simplification of the environment is one of our key objectives, and typically we do that by standardizing on less different types of equipment. We also do that by bringing virtualization, which allows us to better share the same equipment between multiple workloads so we don’t need as many assets, which again is a simplification element -- but also significant to cost reduction.

We also simplify things like the IT management processes and tools, and we have a significant set of services around IT optimization and IT service management that can help simplify the environment. All these technologies like virtualization, automation, and optimized service deployment are significant enablers for consolidation.

Gardner: So we’re not just talking about consolidation in terms of physical consolidation, a location, and numbers of services -- we’re talking about a consolidated datacenter approach that functionally consolidates IT?

Comhaire: Absolutely. That’s how we’ve evolved the thinking on consolidation, and it is also why today we call it "IT" consolidation, not just "server" consolidation. We have evolved over time to a more holistic approach of people, process, and technology. As Peg was saying before, we are significantly looking at the roles of the people.

We’re looking at the processes that are being executed to see if they can be simplified, streamlined, and organized according to industry standards or best practices. Finally, we do the technology part of consolidation exercise, we look at standardization and virtualization of the infrastructure components, and make those work together with less things like configurations of applications. All of these are standardized and provide the value back to the business.

Gardner: This strikes me as not just a high economic impact, but as absolutely necessary ingredients as companies move toward total IT transformation. Because the payoff is a 25 percent to 40 percent reduction in total cost, you set yourself up for an agile multipurpose platform, and are reactive both to your legacy as well as to new types of application approaches, such as SOA.

This is really not an option. If your competitors are going about this and you are not, you will be at a significant disadvantage.

Ofstead: Absolutely. That is why so many companies are moving aggressively toward consolidation. Certainly after emerging acquisitions such as the one HP has had. The more we can do to clean up our applications and make ourselves more cost competitive, the better we will do.

Gardner: Great. Well, thanks very much. I think we’ll need to wrap it up now.

We have been talking about IT Consolidation, its strategic, economic, and future-proofing imperatives. Joining us for this sponsored BriefingsDirect podcast has been Ewald Comhaire, the director of the Global Practice for Infrastructure Services at HP, and Peg Ofstead, worldwide solution lead for HP’s IT Shared Services and IT Consolidation Services.

Thanks very much for joining us today. I think this has been an eye-opening discussion for me, and I hope for our listeners, too.

Ofstead: Thanks, Dana.

Comhaire: Thanks.

Gardner: Thanks for joining us. I would also like to point out to our listeners that if you’d like to learn more about BriefingsDirect B2B informational podcasts or to become a sponsor of this or other B2B podcasts, feel free to contact me, Dana Gardner, at 603-528-2435. This is Dana Gardner, principal analyst at Interarbor Solutions, and you’ve been listening to BriefingsDirect. Come back and listen next time.

Listen to the podcast here.


Podcast sponsor: Hewlett-Packard.

Transcript of Dana Gardner’s BriefingsDirect podcast on IT Consolidation. Copyright Interarbor Solutions, LLC, 2005-2006. All rights reserved.

Tuesday, December 12, 2006

Transcript of BriefingsDirect Podcast on Media Delivery Trends with Akamai CTO Mike Afergan

Edited transcript of BriefingsDirect[TM] podcast with Dana Gardner, recorded Nov. 7, 2006.

Podcast sponsor: Akamai Technologies.

Listen to the podcast here.

Dana Gardner: Hi, this is Dana Gardner, principal analyst at Interarbor Solutions and you're listening to BriefingsDirect. Today a discussion around the massive shift on the Internet to media delivery. There has been an explosion in user-generated content, in media and entertainment companies, as well as vendors of software and software services companies delivering client packages -- an increase in the entire software eco-system.

Mike Afergan is here to help us sort through some of the issues inherent in this avalanche of content -- data packets -- crossing the Internet. Mike is a chief technology officer at Akamai Technologies. Welcome to the show, Mike.

Mike Afergan: Thanks, Dana.

Gardner: We are seeing a different type of content distribution need. As a technology organization that is been working for a number of years at managing this -- both for end-user benefits as well as for those distributing content -- can you help us understand what is different now at the end of 2006 in terms of content, large files, and objects? What is different from just a few years ago?

Afergan: Sure, that is a great question. We’ve definitely seen a dramatic transformation over the past several years in terms of what our consumers -- you and I and, of course, businesses -- are doing daily online. Certainly, within the past year or two we had an inflection point in terms of adoption.

A number of different things are driving that. On one hand, you have a number of technology trends. Certainly one of the most significant is broadband adoption, both in the number of households and businesses that are connected online through broadband on a daily basis, as well as in terms of what broadband means to people today.

Business approaches to using this content has powerfully transformed sites from merely a Website to an online extension of their regular business -- if not an online business in itself. For example, in the media space we see several businesses formed around pay-per-content models, ad-supporting models, and syndication models. In the software space we see companies using the online channel as a primary way to deliver their content. The technology challenges get really exciting with the companies that have real online businesses and are enabling that business online.

Gardner: I suppose it was no surprise to forecast five or 10 years ago that we would be in the broadband world. People expected that you were going to give more broadband to more homes and businesses. But I guess what wasn’t anticipated were these new social phenomena in business models. I’m thinking about Web 2.0 companies, social networking, and people sharing massive files, their photos, and high-definition video clips.

Is this where we have caught the businesses by surprise?

Afergan: It caught some people by surprise. Obviously, a few years ago we saw broadband adoption happening at an accelerated rate. Nobody could predict the exact date we would hit that inflection point or exactly how fast it was going to happen. But certainly people saw that it was definitely happening over time, and the question was exactly which model, exactly which approach, exactly which pieces of new technology would really help it to hit that inflection point? Those certainly are the questions we are just beginning to uncover today.

There are many companies with successful online businesses today. For example, there are premium brand websites where you can get an online subscription. There is the CBS March Madness [basketball series], which has an ad-supported base model and had 400,000 simultaneous video streams. But truthfully many companies are still trying to figure out what the most profitable and successful model is going to be.

Take, for example, syndication, which is a very hot topic right now. This is an area where the business models and technologies are still sorting themselves out. So there are constant surprises and a constant evolution here. That is what keeps this exciting, and very interesting -- and explosive.

Gardner: We are seeing the page-view model with advertising being increasingly important. We’re seeing folks like Apple pioneering with a direct sales model with iTunes for music, where you just buy content via a download of a song -- and increasingly movies, video, and television.

We are expecting more of that from Microsoft with its XBox and Media Center. What are the technology requirements now that we are into these different content-delivery business models? I suppose what is common is that more bits are coming down the pipe. Is this an issue of scale? Is it an issue of measuring revenue? Is it an issue of cutting costs? Is there a need for better metrics? What are the technical requirements that are different now from what we forecast just a few years ago?

Afergan: Fortunately and unfortunately it is all of the above. There are many different challenges to realizing online businesses. I talked about pay-for content, for example, or ad-supported content or syndication. The business models themselves are complicated. Realizing them requires overcoming several technical obstacles.

Take, for example, a site like Major League Baseball. There are many challenges, as in any site that wants to have an online pay-per-content model where you are logging in to a subscription. The challenge is to make sure that only the people who pay for the content receive the content.

You need to worry about making sure that certain games are going to be blacked out in certain geographies. You are going to need to worry about controlling and having a rich interactive dynamic website. So there are many challenges around the business applications and logic, let alone the reporting and the monitoring and understanding your traffic, as well as the user events that happen at your site after the events.

Delivery is obviously a big challenge. Another is: How do you deliver the increasingly high-definition video streams, asynchronous transactions, and lots of small images on a user-generated site? How do you deliver all of this content in a reliable, scalable way?

These companies are evolving; they are facing many challenges, both fortunately and unfortunately, in a number of different areas. The real key theme is the logic and the tools it takes to enable that online business and the sophisticated business logic required to have such an online business.

A large part of that is making sure you have the rich interactive high-quality experience for your user because, ultimately, that is what makes it engaging, what makes people come back to your site and makes people click on more pages, more ads, more video, and more information. At the end of the day, that is what allows you to have your online business.

Doing that on the Internet has many challenges. How do you distribute that information through the Internet, which isn't designed fundamentally to handle the notion of a TV broadcast? You can’t show up at a data center and say, “Hi, I’d like to buy 50GB a second of traffic,” let alone thousands of gigabytes a second of traffic.

That does not work, and is not how the Internet was designed. A large part of what we do for our customers at Akamai is to not only give them the tools that enable sophisticated online business logic to do the targeting and the rights management, but also to provide the underlying platform that allows them to do scalable content distribution, which really is impossible using the bare metal of the Internet.

Gardner: One of the other associated trends here is that more and more types of organizations are becoming publishers. Companies and even individuals are becoming media distribution originators. We now have what is known as micromedia companies.

We now have everything from an individual to a Fortune 500 company getting into the act of creating content, distributing content, having a direct relationship either with their other businesses, ecology partners, or end users or other businesses as clients and customers. They do not want to get into the business of inventing the wheel once again for this sort of network services-level activity.

So it certainly sounds like a need for a de facto standard for a platform on the network for doing this?

Afergan: That is definitely a legitimate concern because the other problem that you did not mention is that while they are doing it, all of their competitors are also trying to do that and be more innovative. So our customers come to us to focus on what makes them innovative with their business model. They are looking for a platform that allows them to do that -- not building on the bare metal of the Internet but delivering more functionality with a platform that allows them to worry about enhancing business logic, new partnerships, and so on.

There are a couple of trends here. One is the disassembly of websites. We have many sites that are now syndicating their content out, allowing their videos to be delivered on other video sites, allowing their content to be incorporated in other sites. And to do that requires a couple of different pieces. One is the ability to deliver that content, and another is to have sophisticated business logic that allows you to determine who, what, where, and when your content is available. A lot of customers turn to Akamai for that increase in business logic and personalization that runs on top of our platform.

The other area you touched upon -- user-generated content -- is a massive concern for many customers. The customer needs to have rapid affinity to your site. We see that for the social networkers as well as the brands now trying to incorporate community aspects into their site.

In the past year Akamai has not just enabled the delivery of content from our storage to our servers and to the end users. More and more for our customers we are also handling the upload of end users' content through the Akamai end servers to our storage environment. That allows companies to build explosive, innovative, and interesting new business models without building any infrastructure. The customer wants us to worry not only about the content delivery and what is updated on their site. They also, frankly, want us to manage the content aggregation from the user-generated perspective.

Gardner: I see. So they not only need to worry about everyone simultaneously hitting on their servers to get a download, but as a media company they need to start attracting entertainment, as a social networking company, and so they need to think about everybody showing up at their door at the same time with a 5GB video file, right?

Afergan: Yes, so you need to worry about content heading in both directions. Frankly just the sheer scale of it -- the amount of people coming to you to deliver content, the size of your library and how that grows, and how much content you are going to need to worry about, manage, and control.

Gardner: You mentioned that we are not just talking about the movement or the management of these files; we are really talking about managing the relationships. We are talking about federating this content, aggregating it, deciding who can see it and under which circumstances, and what is freely available.

This involves a platform of some kind for gathering metrics, a billing or a transactional platform of some sort. And that becomes a logic issue, not just an infrastructure issue.

Afergan: Exactly. For more and more of our customers it is about enabling their online businesses. They are coming to Akamai to acquire those tools and logic. With our system any customer gains a configuration that basically specifies their requests, how we handle, how we process, and what logic we run for those various transactions.

Those are the tools that we built up over the years. We essentially provide applications for our customers to build on. It is not enough to simply have content in your site or to put content on other folks' sites. You need to worry about your authentication system, your advertising targeting system, as well as your reporting system. And that is what Akamai is doing for our customers.

An example is in the media space and the commerce space, where there is much more content going around -- but also much more business logic wrapping around that content.

Gardner: When we think about business logic, we usually think about developers. I suppose there is another trend afoot in the marketplace around accessing applications as services, something we call SOA and Web services, and also seeking out infrastructure as services, or services oriented infrastructure (SOI). There are a host of things going on there with virtualization and grid and utility.

Now, without going down that particular grid avenue at this time, what do we have here for developers? Should they be thinking about these issues or should they be going for services that they can access and integrate into their logic that provides a business model, an innovation, for their businesses or their hosting organization?

Afergan: Services are a powerful paradigm and tool that companies and developers can use to develop extensible applications. All the things that I spoke about before in terms of syndicating content and replacing functionality to other sites are generally available through an API-like interface, some sort of SOA.

These architectures are interesting and challenging because they typically flow over the Web, and small connections exchange these pieces of information when there is a request. That is a very challenging type of workload for a Web infrastructure -- where you have packet loss and increased latency, etc.

That information is exposed to different sites through APIs and various SOA or other XML-based or HTTP-type interfaces. So certainly the notion of exposing more of a service is something we are likely to see much more of over the coming years.

Of course, over the public Internet these services definitely have a lot of small HTTP connections, which means that they are chatty and do have to deal with the fragility that is exposed via TCP. There is a challenge from a quality of service perspective that needs to be overcome, but at the same time it does provide for a great deal of functional flexibility.

Gardner: So you need to build an attractive platform of services to address the new business models. And there is also an increasing set of developers using rich applications, AJAX front ends, or who are looking at SOA. Should we expect something from Akamai with a bit more interest to developers?

Afergan: We certainly hope that our service offerings today are very interesting to developers. Our approach has been, and will continue to be, to service the needs of our business customers. Akamai will not be offering developer software per se, but I think you will see us offering functionality that is important for the more sophisticated applications of our customers.

We will also be offering tools and ways for the developers within our customers to better interface with Akamai. At the same time, part of our approach is to be transparent, behind the scenes. We want to allow the customer to write Web services that are most natural for their business, and then allow it to run over the Internet without concern about the Internet quality-of-service, the vagaries of TCP, or what networks are now connected to.

Essentially all they do is make one small change in DNS, work with us to make sure they’re happy with their configuration, and then write the applications just the way they want to.

Gardner: When your customers come to you is there a different technical dialog now than a few years ago? Do you have another level of technical capability that you are bringing to them?

I know you had a lot of software vendors, for example, who are customers, who wanted to distribute their software products, their patches, and updates and so forth over the Internet. You have met those needs, but you are probably dealing with a different type of company now. Tell me a little bit about this sort of technical problem and solution set that the market is demanding right now.

Afergan: Sure. Clearly our customers, just like Akamai, have evolved significantly over the past several years. One of the exciting things for us is to work with these customers on their future projects. So often we’re in situations with our customers thinking about what they’re going to be doing a year or even five years out from the current timeframe. It is exciting for us to see these trends and to hopefully be out in front of these trends by the time they actually are relevant and required as we build up trusted relationships across the industry.

Online businesses are really the key that is driving what our customers are building -- not just putting content online, but putting sophisticated applications online. Our customers are asking us for the ability to have a high level of sophistication in their application, as well as high level of scale.

So not only are they coming to us asking us to move from 100K to 300K to 500K to 1.5MBit streams, but they want to make sure that when they do that, we are going to tie in with their J2EE applications, that in turn is tied into their user authentication system, which in turn is tied into their reporting system, which -- by the way, also has to work with two or three other third-party technologies that they want to support as part of their infrastructure.

It is generally a much more sophisticated application and developer that we’re working with. And what is great for both of us is that we are working with a more sophisticated business model.

Gardner: So you are much more of an ecology hub here, pulling a lot of different parts together to bring about a business solution.

Afergan: Often we won’t be the one pulling all the parties together in terms of a business perspective; that is not our business approach. But often our professional service organization working with the customers will be interacting with the other third parties, and some are already customers of Akamai.

So from the technology perspective, many of those technologies are running on the Akamai platform and working with each other, with us, and with the customer.

Gardner: Maybe you could give us a sense, Mike, of some of the return on investment (ROI) or some of the other metrics of success here. When we are dealing with such vast amounts of content, and people in the business are shifting as they evaluate different approaches to this, what would you tell them is in their best interest in terms of the cost and capability?

Afergan: Sure. We have been able to develop support across a variety of different business models, and have seen many different successes. For example, in the media a well-known Akamai customer such as Apple iTunes fundamentally changed the music landscape. It has over a billion downloads of music, similar to a pay-per-view content approach.

Another example is Major League Baseball. We support over 250 live games per month during the baseball season, with a variety of subscription-based approaches.

Another case is Akamai customer Friendster, in the social networking space. By coming onto the Akamai platform, not only do they save infrastructure cost but they were able to triple their market share in terms of page views -- which in the advertising space, in the social networking world, that immediately translates into revenue for them, with more and more page views.

A great high-tech example is Microsoft who, for their beta release of the Windows Vista software release, did more than 80GB per second of traffic -- one of the largest sustained volumes in the history of the Internet. They were able to run on the Akamai platform without thinking about building out new infrastructure, etc.

So many of our customers come to us and see significant cost savings and bottom line savings. But increasingly our customers are also seeing increased top-line revenue as well through better performance, reliability and scalability.

Gardner: I suppose in this changing landscape people are also interested in risk reduction. I might be in a pay-per-view model today, and I might want to move to a subscription model tomorrow. I probably want to continue advertising. And then there is this whole notion of syndication. How can I position myself to move among these different revenue models from the same sort of platform or infrastructure perspective?

Afergan: Exactly. I mean many of these customers want to have the flexibility and agility to expand really different business models, to be able to move and incorporate different types. Akamai takes that issue off the table in savings, which gives them the time and the money to invest elsewhere in their business.

With the tools and functionality in our platform, we allow them to shift between the models that you’ve mentioned. Their user model within their applications is supported by Akamai and can support authentication if they want to have a pay-per-view content model -- or targeting, if they decide later they want to do advertising. And, frankly, we have some customers who want to do free trials, too, so moving back and forth between the different models.

So flexibility and agility is fundamental to survival and success in the current exciting, rapidly changing marketplace. What it comes down to for our customers is to have a platform they can build on, both for their application logic and for their content delivery -- one that is not the bare-metal of the Internet. And that is why customers are using Akamai.

Gardner: Very good, thank you for explaining so much. This is clearly a changing landscape, so we will revisit some of these issues, I'm sure.

We have been talking today with Mike Afergan, the Akamai chief technology officer. And Akamai is the sponsor of our podcast today.

We’ve been discussing the explosion of content and media on the Internet, and how network services and shifting business models are requiring a richer set of capabilities from providers like Akamai. Thank you very much for joining us, Mike.

Afergan: Thank you, Dana.

Gardner: This is Dana Gardner, principal analyst at Interarbor Solutions, and you have been listening to a BriefingsDirect podcast. Thanks for listening.

Listen to the podcast here.
Podcast sponsor: Akamai Technologies.

Transcript of Dana Gardner’s BriefingsDirect podcast on Internet media delivery platforms. Copyright Interarbor Solutions, LLC, 2005-2006. All rights reserved.