Wednesday, October 17, 2007

BriefingsDirect SOA Insights Analysts on Citrix, XenSource, HP and Opsware

Edited transcript of weekly BriefingsDirect[TM] SOA Insights Edition podcast, recorded August 13, 2007.

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 Interarbor Solutions at 603-528-2435.

Dana Gardner: Hello, and welcome to the latest BriefingsDirect, SOA Insights Edition, Volume 24, a weekly discussion and dissection of Services Oriented Architecture (SOA) related news and events with a panel of industry analysts and guests.

I’m your host and moderator, Dana Gardner, principal analyst at Interarbor Solutions, and ZDNet blogger. We are joined by a large stable of experts and analysts this week, and this is the week of August 13, 2007. Let me go through our list of contributors. First I would like to welcome Jim Kobielus, principal analyst at Current Analysis.

Jim Kobielus: Good morning, everyone.

Gardner: We are also joined by Neil Macehiter, research director at Macehiter Ward-Dutton in the U.K. Hello, Neil.

Neil Macehiter: Hello, everyone.

Gardner: Making his debut on our show, Dan Kusnetzky, principal analyst and president of the Kusnetzky Group. Hello, Dan.

Dan Kusnetzky: Good day, everyone.

Gardner: A return guest, Brad Shimmin, principal analyst at Current Analysis.

Brad Shimmin: Greetings, everybody.

Gardner: Also, a regular these days, Todd Biske, enterprise architect at Monsanto, formerly with MomentumSI.

Todd Biske: Hi, everybody!

Gardner: And one additional newcomer, JP Morgenthal, CEO of Avorcor. Welcome to the show.

JP Morgenthal: Hello, everyone.

Gardner: And last, but not least, Tony Baer, principal of onStrategies. Thanks for joining, Tony.

Tony Baer: Good morning, Dana and everybody.

Gardner: We want to jump into two major topics this week, and they are both a result of mergers and acquisitions. First, we're going to discuss something that has further highlighted a hot area, virtualization. That is the acquisition by Citrix of XenSource. We're going to look at virtualization as a result of this $500 million deal at a very high multiple. We're going to look at the implications for Microsoft, for open source, and for virtualization in general.

We're also going to step back a few weeks into July and discuss the $1.6-billion acquisition of Opsware by HP. This is a case where we are looking at management, and an acquisition that might help redefine or expand the role of management, automation, and operations, and perhaps have an impact on SOA as well.

First, let's get the low down on the XenSource and Citrix deal. Why don’t you give us a quick recap on that Tony Baer?

Baer: Well, Citrix has made roughly a $500-million offer to buy XenSource, which is itself a company that was formed around commercializing the Xen hypervisor open-source technology. In fact, it only morphed into a real commercial company within the past 12 months or so, maybe even less.

The interesting thing about XenSource is that it’s been considered to be the leading, emerging alternative to VMware. It essentially virtualizes the machine to a slightly more native approach than VMware. It's a very interesting acquisition because Xen has had a relationship with Microsoft, where it gets access to Microsoft's virtualization technology, and it also fills a key gap for Citrix.

Dana, you and several of the other folks here who were on the call raised a lot of questions in terms of how this impacts the relationship between both companies and Microsoft.

My quick take on it is that in the grand scheme of things, it's not going to make huge changes. Of course, there's a lot of wiggle room here, but my sense is that, when all the debris settles, Microsoft will still need some way of interoperating its hypervisor with the Linux environment. So, even though the relationships may change somewhat in the long run, there will still be some sort of technology sharing here.

Gardner: Let's go to Neil Macehiter. The question for you is how big a deal is this really, and is it a game-changing event, given the high multiple?

Macehiter: I'm not sure one can necessarily equate the high multiple with its being a game-changing event. Clearly, XenSource has about $8 million revenues. We think its paid revenue is $20 million next year. So, it's a monstrous multiple, which indicates that virtualization is very hard, as we have seen with the VMware's IPO. They are now the top publicly traded software company in the world, based on market cap. What this acquisition emphasizes, more than anything else, is where the temperature is rising, and it's not in the core hypervisor.

The hypervisor that Citrix has acquired in Xen is actually open source, and the commitment is to retain that as an open-source initiative. The real value in this acquisition is XenEnterprise v4, which addresses some of the lifecycle management issues around virtual machine infrastructure provisioning, image management, etc. That’s where the temperature is already rising, and it was an acknowledged gap in the Citrix portfolio around desktop virtualization. So, I think it is indicative that virtualization management is hard. I must admit I was quite surprised that Citrix was the acquirer. I was thinking that XenSource was an acquisition target, but I was thinking more of the OS providers, the Red Hats or Novells, acquiring the capability, particularly around the management.

The Microsoft relationship isn't game changing, primarily because Microsoft’s hypervisor technology and the management around it, in Viridian and Virtual Machine Manager, is still a year or so away. It's going to be very interesting to see how the relationship morphs, as Microsoft becomes less dependent by virtue of having it's own hypervisor within Windows Server. Microsoft Research in Cambridge, just down the road for me, was involved in some of the early development at Xen, because that came out of Cambridge university.

Gardner: Okay, Dan Kusnetzky, you write the Virtual Man blog on ZDNet, and you have been covering virtualization for a long time. As a former IDC Group head, tell us a little bit about your perspective on this. From what we saw, from quotes out of Citrix, this deal had been under discussion for a long time. The fact that it came within days of the VMware IPO, which had very great reception, doesn’t seem to be related to this. The timing doesn’t seem to be related, and is somewhat of a coincidence. What's your perspective on this virtualization market?

Kusnetzky: First, I want to expand the notion of virtualization, and to point out it's not anything that’s really new. The model of virtualization that I have been using to examine the market includes seven different layers of technology, each moving some part of an application environment from living in a physical world, to move in a logical or virtual world. If we look at Citrix's strengths, they've always played very well in access virtualization, that is, making it possible for an application to display on a remote device without the application having to know a great deal about what the remote device looks like, the network that it's on, or anything.

They've also got some form of application virtualization, allowing applications to be accessed regardless of where they are, what operating systems they're running on, and the like. They’ve never had a very strong processing virtualization, storage virtualization, network virtualization, security story for the virtualized environment, or management of physical and virtual resources.

If we look at just the idea of what XenSource was doing with their processing virtualization management security, particularly their recent announcement of partnership with Symantec for the Veritas Storage Foundation software to be included in XenEnterprise, you could see that Citrix starts to have a more top-to-bottom virtualization story than they every had before. So, from a product portfolio view, this acquisition appears to make some sense.

From another point of view, though, I have some pretty serious doubts. If we look at the history of technology company acquisitions in the past, corporate culture and management style makes a very big difference in retention of key developers, key architects, and key strategists.

We don’t have to look too far to see some pretty major disasters where a big company acquired a smaller company that had a different approach to business and the key designers left. You could look at IBM acquiring Rohm to get into the telecommunications market, and very short order, all of the people who had made Rohm what it was, went to Northern Telecom. Another example would be Computer Associates acquiring Ingress, and the significant designers and architects moving to Sybase and Oracle, leaving CA with a shell of a company.

Gardner: Alright. So, given that there is some inherent risk in mergers and acquisitions generally, there doesn’t seem to be any obvious cultural symmetry or alignment between these two companies.

Kusnetzky: There are even strong differences. Citrix has been married to Microsoft for years, even though it had technology that allowed it to work with Unix, Solaris in particular, and Linux. They really didn’t focus on that because it would anger the folks in Redmond.

Gardner: Right, and they have managed that relationship very well. It’s been something that I was thinking would not last very long, but it has. Nonetheless, so what gives with this deal? If there are these risks, this lack of symmetry culturally, this high multiple, is this a desperation move, and why the big land grab in the case of this virtualization technology?

Kusnetzky: If we look at Citrix's portfolio, every single piece, service, or product offering is matched by something Microsoft is pushing now. That, in essence, means that Microsoft is trying to acquire the business that Citrix has and slowly remove Citrix from the limelight and off to the sidelines.

They have their own presentation services in the form of terminal services. They have their own clustering. They have their own Go-to-Meeting-like software with the acquisition of another company’s business. Go-to-Meeting was Citrix and Live Meeting is Microsoft. If you go down the list, Citrix is being increasingly pushed off into the corner, and they needed to do something that allowed them to come back and be focused. They needed a broader strategy, one that wasn’t focused solely on access mechanisms. The acquisition of XenSource gives them a broader story.

Gardner: Is Citrix drawing a line in the sand and saying finally, "Okay Microsoft, we played this competition thing for a long time. We've been the little sucker fish on the side of the big shark for a long time, but now we're going open source. We are going to use that strategically, and we are going to go off on our own and try to make real hay on this virtualization thing."

Kusnetzky: That is, at least, a possibility. The majority of Citrix's business is centered on Windows, so they can’t go too far along the path of angering Microsoft, because it would become very difficult for them to get access to key technologies that allow their product to work.

Gardner: Let’s look for a little level set with some practitioners in the field. First, Todd Biske, how interesting is virtualization to you as an architect? Is this something you see off in the future? Is this something you look only for operational efficiency or do you see this as something strategic and perhaps related to SOA?

Biske: Virtualization is definitely something that organizations are looking at right now. For the clients I've worked with, it's been a mix. Some are really trying to embrace it on the server side and make use of it right now. Others are looking at possibly using it on the desktop for developers, when they need to get a specific development environment, but it’s definitely in people’s minds today. So, I would definitely classify it as in the list of strategic initiatives that companies are looking at and determining how to use appropriately.

This is a really interesting acquisition that will help XenSource at least get more mind share in the enterprise. Companies obviously have lots of Microsoft investments on all their desktops. There's a good chance that major enterprises have significant investments in Citrix as well, if they've got any need for remote access for their systems, terminal services, etc. It will open it up to a few more environments to add in this virtualization capability for organizations that were still unsure about what to do with open source. It’s a good thing from that perspective.

Gardner: Citrix is in a lot of major accounts, they're well entrenched, they have a sales force, and they have a presence.

Biske: Absolutely. I compare it to IONA’s acquisition of LogicBlaze. For companies that were looking at open source, having IONA behind that was a good thing, because they've now got a strong name, somebody they can go to for the support around that. Now, you've got the same situation with XenSource. When you have these smaller companies trying to commercialize open source, they still haven’t built the reputation of these larger vendors. So, overall for XenSource it’s a good thing. We’ll have to see whether it really enables them to compete even more with VMware and Microsoft’s coming solution in the space.

Gardner: Let’s go to another practitioner. , tell us a little bit about your company and then quickly how you see this virtualization trend appearing in your client base?

Morgenthal: Thanks. Avorcor is delivering what we call the supply chain as a service. We're really involved heavily in a vertical market SOA around supply-chain management. We've been doing a lot with taking SOA and driving it out in a vertical nature, not just horizontal, where I see a lot of the early work being done. Because of what we're doing, we are working with clients who are facing these issues.

I’ll give you an example where the state of art is, where the state of the decision making is. We have a client that is looking to make a very large enterprise application decision. The company from which they're buying their enterprise application software refuses to commit to virtualization or running on a virtualization architecture. Even though there has been significant evidence of enhanced performance that are running natively on a flat Windows platform in a virtual machine, this company would not commit to running in production in a virtualized architecture. On the other hand, we find that we’d perform better in a virtualized architecture.

We've been dealing with issues of the Microsoft platform for a long time around resource management, where we're fighting with SQL server and other applications or resources, and each one has different memory requirements. This virtualized environment allows us to focus on giving our application 100 percent of the resources, and thereby never running out of things like TCP/IP sockets or having memory thrashing errors slowing down wireless communications, which is critical to Web services doing their job. So, it’s having a profound effect on the production environment.

Gardner: Is virtualization making the operations and architects an offer that they can’t refuse, and is Microsoft going to have to adjust to this reality?

Morgenthal: From a development perspective, that's a given. I, myself, have done what no one expected. I'm actually running on a Macintosh, and that's because I run all my development from my Windows platform stuff in a parallel virtual machine. The reason I do that is because the the MacBook Pro is one of the best Intel machines on the market today. It’s solid. It’s stable. The applications and everything else I do using the Web in my job, run perfectly in that environment, and when I need to do development. So, from a development perspective, it’s a given that people are moving in this direction in busloads.

Gardner: How about the datacenter side?

Morgenthal: On the datacenter side, the promise there is better utilization of resources. As I said, if you really want to get into it, you could find a way to tune Microsoft, but I have a sys admin working in one of my clients who is fighting with this 3GB initialization parameter. When he puts it in one way, one app gets hit when he puts in there. When he doesn’t put it in, other apps get hit, but mine works fine.

This is a clear case of where you go out and get an additional operating system license and put this into each application and in its own VM running on a four-way or eight-way Intel Duo Core 2 machine -- they are running an SAN -- and you have one of the cleanest, most high performing environments I've ever seen.

Gardner: Let's go to Brad Shimmin. Brad, were talking now about the big issue: the high cost of ongoing operations and the interdependency between applications and services within an infrastructure. What’s your take on this virtualization opportunity? Is this really going to be a matter of economics pure and simple?

Shimmin: Absolutely, it's a matter of economics, but not just for the reasons we were talking about. We were mentioning that this is something that helps you better utilize your hardware. This lets you better utilize your software. It basically frees you from having to worry about the stack and the interoperability issues that arise over time in managing a stack.

Gardner: If I can just pause you there and stay on what you just said. SOA is, in effect, heightening this issue, because of the need for discrete services running with their own horses and their own power. What do you think is the relationship between the emergence of virtualization and the accelerating effect that SOA is playing?

Shimmin: You're right there, Dana. There are two levels where this really takes effect. One is, as you were just describing, on a broader level, you need to have interoperability. That interoperability is something that is multilayered and something that you need to be able to standardize over time in a SOA infrastructure and virtualization, once you abstract away all those things that mitigate your ability to do so.

On a more finite level, the part that interests me is the fact that, when you set up a SOA installation, you have a lot of issues you have to resolve initially in terms of, "I'm going to have this version of the operating system, and this version of application server, and this version of USB up the stack."

As we've seen a lot of the vendors, Red Hat and now Novell with their agreement with IBM this week, are trying to say, "Okay, look, we are going to, standardize the stack for you. We're going to tell you that it's definitely something you can count on over time. It's not going to break every time we version a piece of it." What virtualization does is let you set up that verified stack on your server and not have to worry about breaking it down the road, because it's sitting in it's own virtual environment.

Gardner: And doesn’t virtualization also allow you, if you're have a Microsoft shop or you have a lot of Microsoft applications running, to continue to exploit and leverage those investments, but extend them, and without an additional high cost on the infrastructure side of it?

Shimmin: Absolutely. Your lifespan for software is dramatically increased by virtualizing it.

Gardner: Alright. Jim Kobielus, what are we missing here? Is there something that we are not addressing in terms of the impact of 1) this deal, 2) the emphasis that’s placed or the spotlight it's placed on virtualization, and then, 3), how does virtualization affect Microsoft going forward?

Kobielus: I don’t know if we're missing it, but may just overlooking the fact that virtualization is hot this week. Clearly, there have been very important announcements, both XenSource with v4 of its product, the IPO by VMware, and, of course, the acquisition of XenSource by Citrix. That’s hot this week, but really SOA is a virtualization approach.

Virtualization is a long-stroke perspective, as is any approach that further decouples the application logic from the underlying metal. That’s what SOA is all about. Another way of putting that is, abstracting the external calling interface from the underlying implementation of a service or set of services. Virtualization has been around since the dawn of computing. When we stopped programming in machine code and started using FORTRAN, COBOL, as well as compilers and so forth, to put greater distance between the logic and the metal, that’s when virtualization, as an approach, began. It was the big bang of the 1950s and 1960s in computing.

So, in a sense, virtualization is like a cosmic background radiation that still radiates and still manifests itself in complex ways, including, of course, this new focus, this recent focus on hypervisors and so forth.

Cycling back to the importance of this week’s events from Microsoft, what happened this week is good for all three major players. It's good for XenSource, Citrix, and Microsoft. Clearly, it's good for XenSource, because, they've got a tough challenge. They're in a market where their primary competitor, VMware, currently has 85 percent of the market for server virtualization. So, XenSource needed its own mini big bang to give it further momentum to overtake VMware/EMC fairly soon.

Citrix is much larger, more deeply capitalized, richer in R&D and has extensive global sales and marketing, and is a logical suitor for XenSource. It's good for Citrix, because Citrix has several people who have been in the access and presentation of virtualization space for quite some while, and it's well entrenched as a Microsoft partner. So, this allows Citrix, by acquiring XenSource, to put together a complete end-to-end virtualization platform, from storage to processor, to server and access virtualization -- the whole soup to nuts.

Gardner: What about this notion of desktop as a service, where you are not just delivering, as Citrix has done, application interfaces? We're not just talking about the back end, where we are virtualizing instances of a runtime environment, operating system, or stack, but are talking about delivering to end users, the essential ingredients of a operating system desktop with all the associated services and applications that come down in that environment. That's something that might be of great interest to your cable company or telco or someone who is working towards software as a service, but wants to expand on that in to a fuller environment subscription based business model. Does this set Citrix up to move in that direction?

Kobielus: Oh, sure. They've been moving in that direction, of course -- thin-client application virtualization and a multi-client environment focused on mobility. It definitely positions them better. Citrix acquired NetScaler, a year or two ago, and they've got their own MetaFrame Technology. I'm hoping that they will then unify or de-siloize their various virtualization product Citrix upon Xen as their dominant hypervisor, without losing any connectivity or interoperability with Microsoft, on both the access side and on Viridian, as Viridian gets built up.

One of the most powerful things is that Citrix is now the bridge, virtualization-wise, with the open-source community. Clearly, Xen has great traction with Red Hat, Linux SuSE, and, of course, Microsoft, because they have been Microsoft’s primary access virtualization partner for quite sometime.

I want to close the loop on your question, Dana. The XenSource acquisition by Citrix is good for Microsoft, because it allows Microsoft to buy some time until Viridian is ready. A year from now, Microsoft can say, “Oh yeah, we don’t have Viridian ready just yet, but look at this. Two of our primary virtualization partners have gotten together to field an ever more comprehensive virtualization product portfolio, which is integrated or will be integrated fairly tightly with Viridian when that comes out."

So, we'll hear Microsoft saying, “We don’t have it all together today, but we have partners who can give you a fuller virtualization portfolio to compete with EMC/VMware."

Gardner: Hold on. What you're saying is that Microsoft is saying, “We will let our partners go out and to get the beachhead on this business, compete against VMware in the short-term, but then we'll come up and take the whole thing later."

Kobielus: Right. I was surprised that we haven’t seen Microsoft acquire Citrix outright.

Gardner: There's another thing I've seen on the blogosphere is people saying, “This sets Citrix up to be an acquisition target.” It might explain why they went for such a high multiple. They think they can recover some of that from perhaps an acquisition from Microsoft, HP, or maybe even IBM.

Macehiter: Can I just chip in one quick thing here?

Gardner: Certainly. Go ahead.

Macehiter: One interesting element of the acquisition is that historically XenSource has been primarily focused on virtualization in the server infrastructural environment. Citrix has historically focused on virtual desktop, and VMware had it's VDI initiative around that. One question about this is the extent to which the Xen capabilities are harnessed by Citrix to address some limitations of the virtual desktop infrastructure environment, in terms of an integration with things like the Citrix Desktop Server. That becomes the primary focus in terms of offering a virtual desktop infrastructure, the expense of the core backend server infrastructure virtualization.

There's an interesting dynamic there. Although, in principle, they have the soup-to-nuts solution, the question is whether they bundle this as a core capability, which means that they are primarily focusing on virtual desktop, versus virtualization across the page. A key issue is the scalability limitations around things like presentation server, where you just see organizations throwing in more and more servers just to deliver the scalability of presentation server.

Gardner: Let's go back to Tony Baer. Obviously, it's a very complex acquisition in terms of its implications. I think we're still wondering. There's a lot of mud in the water. Microsoft can play this in a number of different ways. Tony, do you have a contrarian view about whether this is good or bad for Microsoft?

Baer: I've also been monitoring a bunch of the blogs, one of the more interesting cases that I have seen is that -- and which actually makes some sense -- is that Citrix is not, at its base, an open-source company. As Dan was talking about before, there are some cultural challenges there. XenSource was an academic project at Cambridge, and was then spun off into a non-rofit foundation like the Eclipse.

That would clear the way for Microsoft to pursue an acquisition here, and make for a nice clean break. I don’t know if that answers your questions, Dana. but my sense here is that this might actually smooth some rough edges around the challenges Microsoft has had in developing Viridian.

Gardner: So, you think that they would maintain a close relationship with Citrix, start begging off some of the technology from the open-source community around XenSource, and use that to bolster its entry into the market, particularly to compete against VMware.

Baer: If you have that clean break, where the open-source stuff is taken care by an outside foundation, that could set some sort of precedent from Microsoft, because Microsoft had been making some initiatives towards the open-source community in the area of interoperability.

Gardner: Microsoft is the only major software company in the world that doesn’t have an open-source benefit that helps it with its R&D, helps it in terms of its entry to market, or it gets community to help it develop around the edges of commercial products. That’s not sustainable, is it?

Baer: Not in the long run, but my sense is that, what Microsoft is trying to do in sort of a life-extension mode is organize these interoperability agreements with folks like JBoss. That type of precedent could happen here with the Xen Technology, and if XenSource under Citrix divests the technology and it's absorbed into an open-source foundation, that would clear the way for Microsoft to open up a much closer relationship with Citrix. Ultimately, Microsoft could acquire them, if Viridian is further down the pike that they think.

Microsoft has had challenges developing this technology over the past few years, and Microsoft hasn't been reluctant in the past to acquire technology. The only thing that might be a game changer is that Microsoft tends to acquire small start-up companies, and Citrix wouldn't fit that mold. But, this does open up some possibilities.

Gardner: Todd Biske, as a practitioner in the field, you're dealing with a lot of environments where there's Windows. There are application costs that are increasing and there’s complexity around SOA creating the need for automation. Do you think that this deal for a buyer in an enterprise makes any sense, and how would it help them?

Biske: If you start getting into the automation space, the HP-Opsware deal is obviously the more interesting one. There's a natural connection between the virtualization space and some of the movements in the management space. Your panel here discussed SOA and virtualization almost a year ago, and I have some comments on my blog about it.

When you really embrace SOA, you're going to wind up with more moving parts for a given solution. And in doing so, you could create this management challenge of how to allocate resources for each of these individual services that have their own life cycle. There's a natural potential to move towards server virtualization to do that, so you can get your arms around that whole management concept. Where I've been disappointed in the management space, however, was that we really haven’t seen anything from the large systems management vendors to start tackling this problem.

Gardner: That’s a great segue, and let's go right into the next subject. We found that SOA will have an accelerating effect on the need for, or the pursuit of, virtualization benefits. The same can probably be said for management, but management is, like virtualization, a large and nebulous topic that encompasses many different things. We’ve talked a lot on this show about the governance and design-time management issues, but how do we start looking towards some automation on the operations side to create some kind of a feedback-loop opportunity for services, how they are used, how they are provisioned, and how they are governed?

So, to this Opsware deal. Opsware was founded and created by Marc Andreessen of Netscape fame. It had an interesting see-saw ride in terms of its market capitalization and then the dot-com boom and bust, hung on through the tough times, and now has been acquired by Hewlett-Packard. Tell us again with a little bit more depth, if you would, Todd, how you see management on the operations side and SOA relating?

Biske: What's most interesting for me is applying SOA to the management space. So, if we are creating lots and lots of services -- you may now have 500 or 1,000 services -- you have to look at that and say, "I have a bigger management problem." There's no reason we can’t take the concepts of SOA and apply them to the management environment.

So, whether it's automated provisioning of solutions, automated policy management, a need to change SOA’s or enable more resources for a particular consumer, there is no reason that I shouldn’t be able to have my management systems call a service to do that. I may want to set up custom orchestrations for how to manage my infrastructure. I may want it all automated out of the box and just push a switch and have it happen.

In order to get there, we've got to have management services on all of our infrastructure, and that’s where there's a huge gap. Everything is still intended to be used by a person. Maybe with some creative scripting, people are able to do it, but you can compare it to the days of Web-enabling mainframes, where the technique was to screen scrape off the green screens. You almost have to do the same thing from the management side now. Look at these user-facing consoles, figure out what glue you can put in front of it to script it, and automate it.

Gardner: Let's go to Brad Shimmin. Brad, are we in an area here where we've got management integration pains, and is HP in a position through acquisitions in it's OpenView legacy to help solve some of that?

Shimmin: They're absolutely in a position to help us with this, both from what Todd was just saying there, and the inverse of what Todd was saying, which I want to talk about, but they have the products and the technologies. The Systinet acquisition that I think is the linchpin here for making this all work together. As you were saying, we’ve seen precious little come out of them and others in the space. The real go-getter here for me is the opposite of what Todd was saying. I would want to see my SOA installations able to speak to and hear from my datacenter systems management solutions. And right now, for that closed loop you were talking about, everything we see is the basic SNMP traps that may get read by Tivoli’s program.

That lets you say, "Okay, there is an alert that one of my servers is overrun on memory. I’ve got these applications running on it, I am going to need to do something, and I see an alert that I can drill down on, and do some basic root-cause analysis." That’s not enough for a true business technology optimization and being able to utilize the resources you are trying to marshal for a SOA environment. I want my Tivoli Application Manager to fully automate that process, look at the variables and the event stream coming from my systems, correlate that, and put it into some sort of context with my applications that are running on it.

Gardner: Morgenthal, in some of the installations that you’ve been working with on SOA, has management already become an issue? That's to say, it’s hard to factor what’s running where and how, and getting a holistic view is next to impossible.

Morgenthal: I had a conversation with Paul Preiss, who heads up IASA, the International Association of Software Architects, about this very thing. He has raised a point and is trying to drive attention towards the exponential growth of SOA, as people start to add services and services become dependent upon services and organizations. I don’t see that problem. One of the differences is that when you come at it from an enterprise IT perspective, and when you come at it from a product perspective, you end up with two different SOAs.

From a product perspective, when you deliver a product as a SOA, you're delivering the architecture, and then you are delivering the implementation of that architecture. Therefore, you have a very controlled instance in which you can manage very easily without needing large governance controls, because you're providing all the infrastructure for management of that SOA as part of your application.

Yes, when you have a lot of legacy infrastructure and legacy investment, and people without the sophisticated SOA architecture experience on staff start developing services, you open yourself up for potential disaster. In those instances, the organization has to ask itself, "Are we ready to invest here?" Every organization obviously wants to take advantage of the latest technologies. This is one of them that can really end up biting them if they are not careful.

So, they need to step back and think about what it takes to invest in SOA and start to wrap their legacy systems and make them available. For one client I worked with -- I love to tell the story, it’s hysterical to me -- we went in, did a big installation, and they knew nothing about Web services. We introduced a lot of the organization to what the Web services can do. Before we knew it, there was a line out the door: "Can I have a Web service. Can I have a Web service?" We had opened up doors to data that had been previously locked tight and made it very difficult for them to do their jobs.

I told the CIO, "You have to think about a security architecture around this," and he agreed. I came back three months later to do a follow-up meeting with him, and I said, "What did you end up choosing?" And they hadn’t. And this is an environment where they don’t even allow you out to your Yahoo Mail. Every Website is locked down tight, but anybody with a notepad, and knows how to write an XML document, can now get whatever data they want out of the mainframe.

So, you see what’s happening. Governance is at the point already where you have a consolidated approach. Prior to that, you need a plan, you need a well thought out approach to what it means to invest in this type of technology and architecture.

Gardner: Neil Macehiter, it looks as if HP is drawing some lines between governance and operational management. Clearly, they’ve got a lot that they can pull together to pull off something like that. In your thinking, is governance more important or is management something we need to address in the context of governance? How does management governance SOA fit in your thinking?

Macehiter: I take a slightly different perspective on the distinction between governance and systems management. I actually think it’s a continuum. Governance is about the way you manage the entire service lifecycle. Historically, there has been this distinction between the governance of design-time processes and the design and development of application solutions, and the governance of the operational environment. Primarily, that’s been because applications have been developed and thrown over the wall.

As we move toward more of a service-oriented approach, the service actually becomes a common element that spans development and operations. That’s one of the key benefits of a service-oriented approach. The implication of that is that you can’t make these broad distinctions. There needs to be continuum. And that’s where things like the registry and the repository become key, because they contain and manage the artifacts that can provide that linkage in terms of things like contracts, and then policies.

There has to be more of a joined perspective. HP, in principle, through its acquisitions, has the assets. For example, when they acquired Talking Blocks that was one the early Web services management players that had assets around management of Web services. Now, they're using it to manage their infrastructure, using Web services. But the Systinet acquisition was part Mercury, which is primarily focused on the design-time. HP really needs to join this up. The other issue is around granularity. When a lot of the management vendors talk about managing a SOA, they're really talking about managing the service-oriented infrastructure, rather than managing the services themselves. So, there is a granularity issue.

Opsware is very good at automating provisioning in the lifecycle, but it’s around the infrastructure that’s running the services, not the services themselves. That’s where the linkage needs to come. I tend to share Todd's views. The vendors in the space -- the BMCs, IBMs, HPs, all of them -- have really missed this, and they’ve been lacking in explaining how they're really going to manage the services, because they're so fine-grained. Historically, managing an instance of an SAP application server is very coarse-grained, and that’s comparatively straightforward. But, when you're talking about disaggregating that and having application components everywhere, you have to disaggregate the way you manage as well.

Gardner: It's a much more complex situation. Tony Baer, in the past, we’ve seen two different directions, when we were looking for a continuum-type benefit. If we're looking for a management and governance benefit around SOA implementations, it seems we can go with one big honking vendor who can do all of this. There are probably only a handful of those. Or, we can look toward standards, and such that many different parts can play together and create a solution approach.

It seems like we’re missing not only, as Neil pointed out, a vendor to step up to the plate and solve this on their own, but we are also certainly missing a management-standards approach that would allow a solution-based, ecology-based, or even an open-source based approach. What do you make of that?

Baer: There’s no question about that. Part of the problem is that at that level, when you start dealing with questions of what defines a service level and a service-level compliance and that starts to get into some higher level areas. When you look at the history of standards, the OSI stack is probably a great example of this with the seven layers. I may get my numbers wrong here, but it was relatively easy to standardize Layers 1 through 4, but when you got to layers 5, 6, or 7, where you get close to the application layer, it gets a lot closer to a lot of the assets that vendors consider to be their value adds.

So, I wouldn’t hold my breath waiting for standards. I have always been concerned about what I wrote about a couple of years back as sort of a blood-brain barrier, which is that you had this conception of runtime governance of SOA, then you had this idea of runtime management, which is essentially more of an infrastructure area. They're handled by two different constituencies, one area being handled by datacenter operators and sys admins, and the other area being handled more by the application folks.

So, when I speak with the AmberPoints of the world they say, "Well, what we deal at service level, we're just dealing with it more in terms of tracking -- is the service level maintained -- but we are not going back to the root cause, which is that a particular data server goes down or something like that." There’s just a huge gap there. I was very disappointed with HP, after they acquired Talking Blocks. I never saw much action there in terms of trying to bring that more into what was then OpenView.

HP has a golden opportunity right now, especially with the way they’ve been running HP software. They reversed the acquisition, where they buy Mercury and then put most of the Mercury execs in charge of strategy, which is a brilliant move. They never had much strategy there before that, but they also were taking management to a much higher level. They need to do provide that sort of unity there. That way that you can get a pass-through, so that, for example, when Mercury Quality Center is recording a problem with the design time, that can trigger a test.

Another area is what I would call IT process automation, which traditionally has been called "run book automation." You might recall Opsware had acquired this company called iConclude, which had that type of product that allows you to automate the workflow of running the data center. This is another way of looking at it. You could call that workflow as a service, and that’s something that is an incredibly open opportunity.

Gardner: And, interestingly, will probably play a huge rule in virtualization, if that scales out.

Baer: Exactly, The other thing that Brad and Neil referred to is the whole idea of governance and governance as a lifecycle. It’s not synonymous with systems management, because that’s just an instance of runtime. There’s the whole design time, change time, and retirement time.

Again, there's a huge possibility for the HPs of the world to knit together a nice federated solution here. There are a lot of opportunities that vendors have not exploited so far, where we could finally start to make runtime governance a real actionable component, as opposed to something in which I just look at a dashboard and then get on the phone to my systems operator.

Gardner: Okay. I hope we haven’t presented more questions than we’ve answered today, but clearly two areas that need a lot of attention over the coming months and years are virtualization and how SOA is a catalyst towards a deeper use of virtualization and more economic benefits from virtualization, and, on the other hand, management, how it relates to governance and how SOA will accelerate the need for more of this continuum benefit. Management on the operations side, governance, services, orchestration, infrastructure of services all need to somehow be manageable.

I’d like to thank our panel for joining. These are topics I’m sure we will be revisiting. We’ve had great insight today from Jim Kobielus, principal analyst of Current Analysis; Neil Macehiter, research director at Macehiter Ward-Dutton; Dan Kusnetzky, principal analyst and president of Kusnetzky Group; Brad Shimmin, principal analyst at Current Analysis; Todd Biske, enterprise architect at MomentumSI; Morgenthal, CEO of Avorcor and Tony Baer, principal at OnStrategies.

I'm Dana Gardner, principal analyst at Interarbor Solutions. Thanks for joining. Come back next time.

Listen to the podcast here.

Produced as a courtesy of Interarbor Solutions: analysis, consulting and rich new-media content production.


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 Interarbor Solutions at 603-528-2435.


Transcript of BriefingsDirect SOA Insights Edition podcast, Vol. 24, on industry mergers and acquisitions. Copyright Interarbor Solutions, LLC, 2005-2007. All rights reserved.

No comments:

Post a Comment