Speech Transcript – Bill Veghte, NetWorld+Interop 2002

Remarks by Bill Veghte
Corporate Vice President,Windows Server Group
NetWorld+Interop 2002
Las Vegas, Nevada
May 8, 2002

BILL VEGHTE: Good morning. Now, did everyone get their Krispy Kremes? I know it’s late mid-morning, but Krispy Kremes are an important part of the diet, and they are also a great Windows Server customer. I know, as the press releases come out this morning, they announced they are moving their whole infrastructure to .NET and to Windows Servers. And what that means to me is that I will get an alert when my Krispy Kreme doughnuts are ready and hot. So, help yourself.

We’re actually very excited about a number of press releases that have gone out over the wire on a couple different themes as there are a couple groups in the audience, press, partners and customers.

And I wanted to do three things. I wanted to talk to you about where we are in the marketplace with Windows 2000, and the momentum we’re seeing there. That’s really exciting in these tough economic times to see that continuing to grow.

The second is, I want to talk about our interoperability strategy. It’s something that we are fully committed to, we understand that even as customers migrate to Windows servers, they have a lot of investment in other platforms, and in particular UNIX, and so understanding the strategy with which you can interoperate and evolve, that is the question that we often get, a scenario that we are investing in, and I want to lay out the strategy.

And then I want to make some, I think, pretty exciting announcements around SFU 3.0 and the relationship and partnership we have with Rogue Wave.

All right. So, with that, let me jump into, this is my favorite slide because it’s about the numbers, and about the momentum we’re getting in the marketplace. Okay, the economic times are tough, IT spending is down, server shipments worldwide are down. Worldwide they’re down by almost 7 percent. In that marketplace, Netware, slightly down; Linux, flat; UNIX going from the high teens to about 12 percent; Sun, down 30 percent on a unit basis; Windows, up. Windows is the only server operating system worldwide that is up significantly for five consecutive quarters. We’ve quadrupled, as IDC is reporting, we’ve quadrupled our install base from 2000 to 2001. And, in the same period of time, we continue to grow our ecosystem and our partnerships. You bet on us as a partner, and you bet on us as a technology supplier on the basis of not only the technology investments we’re making, but in the ecosystem and the growth that we build together.

So, the numbers look good. There was a Morgan Stanley CIO survey that I thought the methodology was very, very good, where they surveyed CIOs in North America, that study


and they were basically ranking the projects that they had very high, which was Web services, which of course is something that is near and dear to our heart; Windows 2000 deployments speaks for itself, and then enterprise application integration.

So, customers are voting with their wallets, and saying that Windows 2000 is a very high priority project, and we thank you for that support and that momentum. So, those are the numbers.

Now, let’s talk about some specific areas and benchmarks that I think are exciting progress. When I talk to customers or partners or press, they say, all right, we believe that you’re getting there. PC economics, we love the value proposition you bring. But you’re not quite there. And so, we constantly push and push on that. And I look at this and I say, I look at the recent Gartner report, and the recent Gartner report said, wow, from reliability and scalability and performance perspective, Windows 2000 is on par today with UNIX platforms. I love that. I look at industry leading benchmarks, such as TPCC, where all eight of those benchmarks and those world records are on the Windows platform, all eight of them. You look at that and you say, not only that, but the top Windows TPCC benchmark is at 709,000 CPMCs, where the top UNIX benchmark is at 455,000. So, not only are we the top eight, which is all they publish, but the top eight, but we blow the UNIX platform away. In partnership with Unisys and the ES-7000, we have two of the top ten non-clustered solutions, and it doesn’t take a rocket scientist to figure out we want to get more of those. And we are confident in the coming year in partnership with some of our key OEMs, and with the delivery of .NET, we’ll keep pushing in the scale up, and in the non-clustered solutions as well.

Now, I look at that and I say, okay, benchmarks are great. What about a typical, let’s say an eight-way Dell or HP solution. You look at that and you say, let’s benchmark comparable performance between say a Sun E-4500 and a Dell or an HP eight-way solution. Comparable in performance at a third of the cost. Okay, so that’s TPCC, and TPC.

PeopleSoft, we hold the world’s record benchmark for PeopleSoft. SAP, three weeks ago


or actually it’s probably about a month ago


a month ago now, we announced that we had set a new world’s record on the SAP three-tier retail benchmarks, set a world’s record on that. And the thing that I’m particularly proud of is, not only did we set the world record with Windows, but we set the world’s record with half the number of processors that Sun had on their previous world record. OK, so keeping benchmarks are a good way of measuring us, and are we pushing up into the highest scale, highest performance segment of the marketplace.

Total cost of ownership is paramount. How do we think about the total cost of ownership of the Windows platform not only in the context of acquisition, but in the context of maintenance, deployment, and operation? Aberdeen did a very comprehensive study over three years with 100-plus customers for medium and enterprise businesses comparing and contrasting the cost of the Windows platform versus UNIX, and demonstrated in their model that there’s a 46 percent savings on the basis of the Windows platform. META Group did a similar study focused specifically on ERP and showed 50 percent savings. Now, these are studies, I encourage you to look at the studies just as a model for thinking about your customers, or your environment, and the cost savings that Windows represents.

The other side of it that when I talk to people that I think we miss sometimes, and the marketplace misses, is also time to market. When I think of the value of the Windows platform it’s, yes, I have to demonstrate that we have very high performance, we have great scale, we have lower total cost of ownership, which is more an operational statement than anything else. But it is also the agility of the platform and your ability to create a competitive advantage on the basis of those technology building blocks. So, I ask myself, and frankly push the development team, are we creating a platform and a set of software building blocks that increase the agility of your solutions. How long does it take for GMAC to process a loan on the basis of our building blocks? How long does it take Home Shopping Network to get their solution up and running on our platform, and then how long does it take them to make changes and tweak their e-commerce site? So, when we think about our value and what we’re all about enabling, it is about that agility that creates the competitive advantage for you.

Those are all nice bullet points. Let’s talk about some customers. If you have a chance, there are a bunch of different press releases coming out today, not only from Microsoft, but also from Unisys, also from Stratus, also from Rogue Wave, and as you read through those press releases, I actually think the most exciting thing about it is some of the customer studies. In doing this presentation, it was hard, you know, like we have Sumitomo Bank and Matsui Bank in Japan, they have merged, and they have two different headquarters. They have a headquarters in Tokyo, and they have a headquarters in Osaka. And I think it’s about 300 kilometers apart. And they are using Windows 2000 Datacenter and the Unisys ES-7000 to do failover, and they have done a failover in about a minute. So, you look at that and you see real world using it today.

You look at the top company is the fourth largest e-commerce site in Australia, and they are the largest florist in Australia. They’re coming up on Mother’s Day, so a big business time for them, and they’re using

‑‑ they have switched from UNIX to Windows 2000 Data Center and Stratus for their entire order system.

You look at the City of Minneapolis, they had a bunch of different server platforms, and they consolidated them on Windows 2000 and the Unisys ES-7000.

You look at South Carolina Blue Cross/Blue Shield, they were on a UNIX platform, they were actually using I-Planet. They have migrated their million users from I-Planet to Active Directory, and are using Active Directory for their Internet solution.

And you can go on and on. I look at Home Shopping Network, and I get excited about how long it took them from going to concept to using Windows online as their primary business solution.

I look at GMAC, that was a very, very competitive bid for us, competing against Sun. And in that situation not only were they able to get this Windows solution up and running very, very quickly, but they’ve been able to take their loan turn around time, their loan turn around time was averaging anywhere from 60 to 90 days, now they’re down in the 10- to 15-day range. So, I think about that as, boy, we offer the economics and the approachability of Windows, but we also offer greater agility.

When you look at the press releases, and you go to use our Web site, there’s some great customer case studies out there. Look at how they’ve done things. If you don’t like the case studies, you need additional content information, push us on it, because case studies are one of the ways that I think we need to speak to you and to the marketplace to demonstrate it isn’t us saying, hey, we’ve done it. I mean, it’s important that we show that we can set world records on the Windows platform in the TPCC or PeopleSoft or SAP benchmarks, but at the end of the day, real customers utilizing our platform to create real business value and real competitive advantage. So that’s the move to Windows.

Now, that is what I’ll call the motherhood and apple pie, bread and butter. When I think about what we need to do as a technology supplier, as a partner for you, it’s not only that we can demonstrate a great trajectory around the abilities. It’s not only that we can create better business value and agility. It’s also that we have a vision and a business model that lets us continue to be very aggressive in the innovation and the technology solutions that we enable.

And, as we were thinking about our application platform, and this really goes back probably four or five, probably actually more like five or six years now, it was clear to us that a new development model was going to emerge in the marketplace, and we sat and had a bunch of


actually it was over a couple of months, a set a real tough strategy discussions about how important was cloud base, was there going to be a new paradigm from client to client server to server centric to what we call the cloud? And that ultimately took the form two years ago of .NET Day, and .NET Day was fundamentally about Web services, it was about data interoperability, and it was about identity.

So, creating a vision around Web services at the time it was, I would say, controversial, a big new thing. Today, people take Web services as the direction and the trajectory that the marketplace is going as a given. It’s not only Microsoft that’s saying Web services. It’s companies like IBM, or even Novell is saying, hey, we’ve got a Web services strategy, too, and we’re pushing that ahead.

Web services as the next generation is important, and then what we needed to do is, we needed to create and build a tool that enables the development of these next generation applications and Web services. And we delivered that with Visual Studio .NET. Visual Studio is the tool, it is the industry-leading integrated development environment, IDE, in the marketplace, taking that tool forward and utilizing it to enable the development of Web services and next generation applications. Okay, that’s the tool.

The next thing you need is, you need a server to deploy and operate these things. That server is starting with Windows 2000, and then moving forward to Windows .NET Server. So, now I’ve got a good strong vision that we’re investing heavily on, rallying our development team, our focus, our efforts, and the industry around. We’ve got a server that enables the deployment and operation of it.

The piece that then we get consistency back from customers is, hey, I like this Web services thing. It’s great, that’s where I want to go. But I’ve been spending years and years investing in a UNIX platform. I’ve got a lot of UNIX or legacy applications. Give me a migration and interoperability story. And that’s what we’re here to talk about in the second part of the presentation.

I have this investment and I want to maximize it. I want to be able to amortize it over as long a period of time as possible. Maximize it not only in the application code, but in the expertise and knowledge of my people. I’ve got developers who know how to write these applications. I’ve got people that know how to operate and run them, and I’ve got years and years of scripts. So, maximizing the previous investments, lowering costs, in these tough economic times, the justification, the economic justification is even more important, and people look much harder at what the return on investment is those projects. How long does it take me to do the turnaround? If I put some money in, what’s my payback period?

And then, frankly, it’s about, as I do new projects, as I do investment, you know, in the case of Krispy Kreme, they were thinking about their growth is just exploding in their franchises, they needed to make a long-term bet on what their application development platform is, and what their server platform is. They made that bet on the basis of not only what they saw as great economic justification today, but on the basis of the long-term vision that we were projecting in the marketplace and the R & D investments that we are putting behind that.

I’m responsible for the Windows Server business and group at Microsoft. My teams do all the core R & D development around Windows Server, as well as I run the product marketing organization for Windows Server. That investment, even in these tough economic times, will grow next year. We are continuing to dial up that investment because we have a business model that supports it. And in order for us to be a long-term technology supplier for you, it means that we continue to very aggressively continue the trajectory in server innovation and server value. So, migration requirements, critical.

So, then what happens is, customers say, look, okay, I get it. I’ve got to maximize my current investments, I’ve got to have a very quick turnaround time on any investments I make, and I’ve got to be able to bet on a long-term leader. So then, I look across the taxonomy and the investments I’m making, what do I need to do? Do I stay put on a set of things? And staying put, I talk to customers and partners sometimes, and people think staying put is a no cost option. Staying put, there is ongoing investment. It’s like owning a house, right, there’s ongoing investment. You’ve got to paint it. Things wear out, you’ve got to keep it updated. The marketplace changes around it. And so, you know, you want a new sidewalk, or in my case I wanted to put high speed networking in my house, and I had to rewire, and all this and that. But staying put is an ongoing investment. Don’t be confused about that.

Then there is interoperability. So, it says, my next bet is on Web services, Microsoft or somebody else’s, but I need to be able to interoperate very smoothly between the stuff that I’ve gotten invested in over the last 15, 20, 30 years, and the things that I’m doing going forward. Slightly higher cost, but hopefully on the basis of the next generation things offer payback.

Then, there is migration. And migration says, I want the PC economics of the Windows and Intel platform. I want those lower costs today, and so I’m going to migrate a specific set of things. It doesn’t mean that I’m giving up on interoperability in some areas of my IT investment, but I’m going to migrate.

And then the last piece of it is evolution where I move it, and I bet the farm on a next generation application development model, next generation enterprise application integration.

So, when I talk to customers, it is about, what are the areas we want to stay put on, what are the areas we want to interoperate on, what are the areas we want to migrate, and what are the things that are so long in the future that frankly we just want to throw out any ball? And so, as a result, when we go through that analysis, it is very clear that interoperability, migration and evolution are pretty fundamental to our success as a technology enabler, for businesses.

And as a result it’s an area that, frankly, we’ve invested in for many years though I don’t think the marketplace understands, or certainly the perception of Microsoft as an interoperable vendor is not


the perception is lower than what we’re delivering on. There is no commercial enterprise server platform in the world that is investing as much as we are in interoperability. It starts in the core, which says, does our TCP/IP stack, or our PKI implementation, or HTTP or XML, are those things designed and developed to be interoperable? Absolutely. Absolutely.

Now, the next thing is, those things, HTTP is the basic transport; XML is the protocol for data interoperability; those are sort of basic elements, key foundational pieces. But they’re not enough. It is about those pieces in conjunction with, and now I’m going to the outer part, a set of solutions that enable application interoperability, that enable data interoperability, or network and management interoperability. So, as a result, let’s take management in the directory. We spend a bunch of time thinking about how we can migrate, but also interoperate with I-Planet, for example. Some of you may be familiar with a product that we have in the market and ongoing development called MMS, and the basic premise between MMS is that it is a pump back and forth between Active Directory and other directories. I’m over-simplifying a little bit, but so that you have very good interoperability between Active Directory and other directories. You have services for NetWare.

And then you have a variety of client interoperability pieces. We have great interoperability, of course, with Windows clients, but also with Macintosh clients, with UNIX clients, and frankly with not only PC clients, but device clients. We spend a bunch of time thinking about how do we interact, and what is the development model, and what is the data interoperability level with PDAs, or with Smart Phones, because I personally believe there is a set of smart clients that augment the PC where data interoperability to the server is super important.

Look at, for example, the partnership that Microsoft has with Symbol Technologies, the leading vendor in bar-code scanners in the world. They have utilized a bunch of pocket PC technology. They deliver those pocket PCs. Companies like Vail Ski Resorts has the Symbol

bar-code scanner entering a bunch of data, and going very quickly against a Windows 2000 Server to get that data real-time in to do credit card transactions to see where the lifts are busiest, and where the lift lines are occurring.

So, we have interoperability with a wide variety of clients, PC or otherwise. It’s about data interoperability. So baking in things like ODBC or OLEDB into the platform. And then what we’re here to talk about today, which I really think is the cornerstone interoperability for the server is Services For UNIX, or SFU, and the Interix Toolkit. Super, super important. It is very clear as we push ahead with the .NET vision that the interoperability and the migration and the evolution of UNIX applications and services to the Windows platform means good interoperability. I listed a bunch of customers on the previous two slides ago, in the press release there are many more that are moving from the Windows platform


excuse me, moving from the UNIX platform to the Windows platform, but it does mean that they still have investments in UNIX and that interoperability is paramount.

We started that investment several years ago. We accelerated that acquisition, or we accelerated that innovation and that investment with the acquisition of the Interix product, and we are here today to announce SFU 3.0. And the premise behind it, and the design principle, are we really enabling the optimization of those previous investments, previous investments from an application perspective, scripts, and experience and skill sets? Are we lowering the TCO so that people can actually run a set of UNIX applications on top of the Windows platform and enjoy the PC economics that Windows and Intel bring to the server equation? And then, do we think of it as a core integrated part of our product line, so it’s baked into how we think about and how we talk about the Windows Server family? And the answer to all those is yes.

But, actually, the person I would like to have speak to it is the guy sitting to my left, Doug Miller. Doug came to us as part of

‑‑ he was CEO of the company that did the Interix product, and he is probably one of the strongest UNIX heads in Microsoft, he is the architect of our UNIX strategy, and a vociferous advocate of this interoperability. And I think this is an exciting day for us, and for customers on the basis of the great work that he has done and led across Microsoft.

Doug, take it away.

DOUG MILLER: OK, thanks, Bill.

So, just to give you a view of what we have in services for UNIX, just running down a couple of the top level nuts and bolts, this is a product that runs on Windows 2000, also runs on NT, runs on XP, which is our client operating system today, and of course will run on Windows .NET server when we release that product in the coming months. I should also emphasize, this is both a product for clients, our professional operating systems, as well as our server operating systems. And well talk a little bit about how it would apply to a client scenario as well as a server scenario.

So this is really focused on solving two major pain areas that we see with customers. The first one is, UNIX to Windows interoperability, the second one is migrating from UNIX to Windows. So we look at the interoperability area, a number of features, and I wont have time to go through every single feature that we have in the product, and I invite you to come back at 1:00. James Slonsky, who is our product manager for services for UNIX, will be doing a detailed rundown on the product. But, essentially you can break it out into two general areas. First is file and data sharing, and here we fully support the NFS protocol, which is a native protocol on UNIX systems for sharing files. We have the ability to, from clients go to UNIX servers and interact with the files on the UNIX servers, and from UNIX clients they can get to Windows servers and be able to, obviously, interoperate at that level as well. And then we include a gateway facility so that, for instance, you can have a network of Windows systems that can seamlessly get access to a UNIX network through a gateway server thats set up in your enterprise.

Next, we have a number of facilities around cross-platform management. And we have some really cool stuff here that allows you to, essentially, if you like, consolidate your directory service onto Active Directory, and essentially expose Active Directory back to your UNIX network and make it look like to the UNIX network its a native NIS directory service. UNIX systems typically use NIS, sometimes they use local password accounts. This is a very nice feature for people to essentially seamlessly allow their users to be able to authenticate off of Active Directory and not have to go through the pain and agony of maintaining double directories and passwords and things like that.

However, if the customer is interested in maintaining dual directories, we support that, too. So you could, for instance, have a UNIX directory with NIS, Active Directory on your Windows system, and provide mapping between the two directories and automatic synchronization of passwords. So lots of options for customers to be able to really set this up any way they like, but make it much, much easier to manage these two diverse platforms.

Next, we have a full suite of tools that are included with the Services for UNIX product. UNIX typically has 300-400 command line utilities, weve included all of those standard utilities with the services for UNIX products. So if you like to do AWK, and GREP, and SED, and LS, and PS, and all that other good stuff, its all there. And works seamlessly with the other utilities built into the Windows operating system. So, for example, you could do something like a DIR, which is a typical Windows or previously DOS command, and you could pipe that to, say, the UNIX GREP command. So lots of capabilities there to build scripts, and enhance existing scripts to take advantage of administering your Windows systems, just like you did with your UNIX systems.

Also, provide remote capabilities, so you can remotely get to your Windows system, just like they were UNIX systems. So things like our log in, our exec, Telenet and that sort of thing are all provided, even provide x-Windows capabilities, so you could have, for instance, an x-term console from your Windows box displayed back, maybe, on a UNIX box that you have maybe in your data center.

The other pain area that we saw is around migration. And now with services for UNIX weve included the Interix technology into a single package. Here we provide a full set of UNIX operating system development tools. So its not just a shell scripting environment, we also have compilers, we have C compilers, C++ compiler, FORTRAN compiler, we also include the standard build utilities…all of those good things, so essentially provide a fully consistent environment for UNIX developers to be able to bring their code to Windows, and be able to recompile that code and have it run natively on the Windows operating system.

So a quick summary of what were doing with services for UNIX thats new with this particular release, I mentioned weve combined the previous two products we had in the UNIX space, we had Services for UNIX in the 2x range, and we had Interix, the 2x range, as well. Weve combined those into a single, integrated package where a customer can install as much or as little of this UNIX technology as they need. Weve made a lot of progress on making this very seamlessly integrated into the Windows operating system. And well be doing a demo in a second here, but whether youre trying to get to files on a UNIX box, or whether youre trying to, say, run a Windows command inside a UNIX shell script, essentially we offer the ability to mix and match literally up to any extent that you feel is necessary.

Next, we have better manageability. I mentioned things like our ability to not only host UNIX directories on the Windows platform, but lots of options for doing interoperability with existing UNIX directories that might be out on your network. Another area were very proud of is around scalability and performance. In fact, just yesterday ZD Labs published a new benchmark. We asked them to benchmark services for UNIX, our UNIX protocol services through NFS for file sharing, and benchmark it against Linux. Take a Linux box, exactly the same hardware, and run a similar benchmark on Windows, similar benchmark on Linux, and were proud to announce the results are essentially the same. And thats always been our goal, which is we should appear to a UNIX person to be having at least as good performance, especially in file sharing, as you would get, say, on a Sun server, or another UNIX based server. So weve done a lot of work there to increase performance of things like NIS and NFS.

DOUG MILLER: And it goes without saying that on the SYS were much better.

JAMES SLONSKY: Yes. So obviously in our native protocols we do very well, especially against UNIX or Linux. But, being able to essentially emulate a UNIX server for file sharing, and being able to do as well as a comparable UNIX server we felt was a great milestone to achieve.

Then finally, great news here, weve combined these products and lowered the price. The combined product is now available for $99 suggested list price, and thats regardless of whether you put it on a server or on a client.

So were going to jump into a little demo here, Ive been doing a lot of talking. So what we want to do is run through, essentially, an application running on a Sun box today, and some shell scripts running on a Sun box, were going to move that application over to a Windows box, and we essentially will move the files using the NFS capabilities and services for UNIX. Were going to recompile that application, essentially without changing any lines of the code, recompile that and have it run natively on our Windows box, but it doesnt stop there. Typically we hear people want to extend the application and take it beyond what it used to do on UNIX, and take advantage of new Windows features. So were actually going to migrate this application and evolve it to become a .NET Web service, and do all of this in a couple of minutes.

So let me jump right in. Ive got a Sun box here. So 32 bit versus 64 bit, so about 75 percent of Sun systems today are still running in 32 bit mode, some in 64 bit mode. But, one of the things that weve done with Windows, of course, is we now have a 64 bit version of Windows. So we have a number of techniques, and I wont have a chance to go into it right now, but I can talk to you afterwards, about how you can take a 64 bit application and migrate that over to Windows, as well.

So let me just jump straight into the demo here. What were looking at, Ive actually got Hummingbirds Exceed X-server running on my Windows box, and remotely connected to this Sun box over here, and old Sun server. So a typical UNIX box here, Ive got a shell open, and we can


so a typical UNIX box, we have a KORN shell open here, and we have a program that Ive brought over, its called X Tide, and essentially this is a program that gives you tide tables for virtually any place on the planet. And weve shown this in some of our workshops, as well. It runs in both character based mode, as well as X-Windows mode. So just to show you that were actually building from the same source code, Im actually going to unpack the source code and were going to build this application on the Sun box, so you can see that it actually is built from an original set of source code. So typically we would run a command called x make mf, it builds you a make file for this particular set of source code. And then we would go through, do a make clean just to make sure that we dont have any left over files from any previous versions of this build, and then were going to go ahead and just build the application.

So one thing Id point out here, just notice how fast this is building, its taking a little bit of time, it wont be too long. But, well be building the same application over on the Windows side, and I think youll be surprised at how quickly it builds from the same set of source code over on the Windows side. So essentially this is using the GCC compiler, a popular compiler in the UNIX world, and in fact, the same compiler that we have in the Interix development environment.

So weve built this application. We can run it in two modes, we can run it in two modes, we can run it as a character based application, and there we can actually give it a location, in this case the default is Golden Gate Bridge, and it gives you the tide tables there, and then we can also run it in an x-Windows version so we can actually display this as a graph. So it seems like a fairly simple application, but its actually fairly complex on the back end, it has a large database that it interacts with to be able to gather this information, and gives you the information in real time.

So enough of the Sun box. Were going to log out of the Sun box now. And what I want to do is go and grab this, and bring this application over to Windows. So to do that Im going to go to My Network Places here, and youll notice that weve added a new network place category on this screen, we now have NFS network as a category for us to go into. And here we have a LAN, called UNIX LAN, and I can go in there and I can actually see Im, in fact, running on my Windows box NFS server, but we want to get over to the Sun box, and the Sun box is called Thor. So this will actually allow us to go under the files on this Sun box and be able to take those files and bring them over.

So Ive got an export here that has all of our home directories, and I can go into my directory where I had these two files. Im going to copy two things, one is a shell script, and the other is the source code itself. Im going to copy those files, and then Im going to go over to a local drive, and Im going to essentially go and drag and drop these files over to the Windows box. So for a Windows user this is a very natural thing to do. I didnt have to know any UNIX commands to get to that UNIX box, and now I have the UNIX source code and my shell script over here.

The next thing I want to show you is the Interix environment, and essentially this is a full UNIX KORN shell that we have running on the Windows box. Were no longer connected t the Sun box, and we dont even need this Sun box anymore, so Im going to unplug it. So now weve relay got all the UNIX we really need, and its all running on Windows. So here we can do any of our favorite UNIX commands, we can do things like LS, and PWD, and PS, and we can do AWK and things like that. So one of the things that we have is the ability essentially to run the same command we ran on the Sun box. Were going to unpack the source code from the X-type program, the same source code we had on the Sun box, go into the same directory structure here. Do the same command, which is x make mf to build a make file, and then were going to do a make clean, make sure that weve got a clean environment, and go and build this application.

You notice how quickly its going, as opposed to how long it took to build it on the Sun box. So certainly no performance penalty for running your UNIX apps on top of Windows. Now we can go off and run the same commands, we can run Tide, and we can even run x-Tide, which is the x-Windows version of that application, now running on our Windows box.

So this is good. One of the things that I did here is build a shell script, and thats the shell script I brought over from the Sun box, that actually goes out and counts the number of lines of code inside the source codes that Ive brought over. So if I look at this, this is sort of a convoluted shell script, goes and finds different file names and types it to WC to do a word count, and GREPs some stuff, and runs it through lock. And essentially if we run this it will go off and tell us that we have 8,000 lines of source code that we brought over from the Sun box.

Now, keep in mind, so far I havent opened an editor to change any code, Ive essentially brought the code exactly as it was over to Windows, and its now running as a native Windows application. So thats not good enough, because a lot of customers say, thats great, now Ive got that same old, boring UNIX application now running on Windows, but the reason for going to Windows is I want to take advantage of native facilities on Windows. So one of the areas that obviously customers are very excited about is .NET.

So what Im going to do is take this application and now essentially wrap it to turn it into a .NET service. So what were going to do here is go off and run Visual Studio. And I wanted to show you, essentially, how complicated it was to wrap this service. We have a project called Final Tide, and really the guts of Final Tide is two C# components that weve written. One of the components essentially wraps any UNIX command that we can run in the Interix environment, and then the other component essentially is custom tuned to the tide environment. So heres our first C-Sharp application, and essentially the purpose of this application is to be able to take a command line application, call that with any parameters that you pass that application, capture the output, and then display that output as XML data. And you can see its a reasonably short program here, a lot of this is auto-generated code. And if we go over to the particular service file that we have for the tide application, and we publish this through SOAP, we then have a way to essentially run the tide command here, and get a list of places we know about, and then also go out and run the tide command with a location as an argument and be able to get a tide table.

This is all great, weve got this


we can essentially compile this in the individual studio environment. And lets test this our and see how well it works. So essentially what Ive done is now advertised these two services. I have one which is list location, so this will go out and essentially ask tide, what locations can you give me that start with a particular set of letters. So whats your favorite seaside resort in the world, anywhere?

DOUG MILLER: Gloucester.

JAMES SLONSKY: Okay. Im juts going to type in the first few letters of Gloucester here, and then well go out and invoke it. And what it does is its going out and calling that UNIX application that we brought over and displaying the results of that run in XML data, and here we have two results that came back, Gloucester Harbor, and then Daytona Beach Shores Sun Glow. It caught the Glow in Sun Glow and returned that as well. So we essentially have displayed this as XML data. I just want to copy this text, were going to go back and run the other service that we build, which is okay, get a printed tide table, so I can go and paste in that location name, and then go and invoke this again. And essentially here weve taken the tide data and represented it as XML data.

So what weve done is built two services that are now ready to be used, theyre not terribly useful as they are right now. So what Ive done is built a little Web page. And imagine that you were a boat rental agency and you had this facility for all of your sub agents to be able to go and get tide data from anywhere in the world, and you want this published as a Web service, out on the net, thats discoverable through SOAP, and that they could go and use that any way they wanted to, independent of where its located. So here we can go in and type in Glo like we did before, and essentially this will go out and look for all the places that have Glo in it, and we found the same two again, Gloucester Harbor, and I can hit okay, and what it does is then go and return a tide table.

Not terribly exciting, but it at least gives you a good example of how, with very little investment of additional technology, were able to wrap a legacy UNIX application and bring it forward into the .NET world.

DOUG MILLER: Okay. Thats great. Lets get back to our presentation here.

So the next thing Id like to quickly talk about is great services for UNIX. This is a product typically customers are asking us, we need solutions. And so I really want to frame this in what were doing around UNIX migration and interoperability throughout Microsoft, and how were working very closely with partners to deliver solutions to the market. This is really focused in two areas, technology, so obviously youve seen Services for UNIX, but were also making a number of improvements to Visual Studio to make it much more approachable for UNIX folks. Were also improving things in Windows.

For example, we now have command line utilities for virtually every administrative function that you can do on Windows available from the standard command.exe command line, also looking at investing in a new shell for Windows that will be, again, much more approachable for UNIX folks. And probably more importantly, were working very closely with third parties who have solutions in the market today. So were going to be talking about Rogue Wave in a second here. Were announced a relationship with them, but we also have press releases here, announcing new relationships with F-Secure, who have some SSH technology, Hummingbird, who are in the audience here today, a market leader in X-Servers. Accu Corps, who have developed a number of COBOL solutions that run on top of services for UNIX, for customers migrating COBOL applications to Windows.

So this is really just the tip of the iceberg, were looking at building these relationships with technology companies, and then also focusing on services. And this is an area customers have given us a lot of feedback on, which is dont just give us technology, help us, give us some content, give us training, give us prescriptive architecture guides. And so were building a set of technology offerings around content, as well, and then also building service offerings for things like doing assessments and doing deployment planning and that sort of thing, as well as the actual migration services.

And again, super critical, were working with partners here. Were also announcing today a relationship with MigraTec. MigraTec have a tool for doing assessment of UNIX code, and being able to generate a report of how much work it is to actually do a migration. Theyre in the audience here today, as well. We also have been working with people like Infosys, Interop Systems, TPI, Boston Group, and others, to provide services to customers around migration.

And finally, the last thing Id like to point out, we launched a new portal a couple of weeks ago, and this is a portal around why customers should migrate from UNIX to Windows, as well as how they can migrate around four different scenarios. I actually just had some interesting news back from a company that measures business portals on the Internet, and its the fifth most popular portal in the world right now.

JAMES SLONSKY: If you have a chance, do use this portal. These guys have done a very good job on it, in terms of how to think about the value and the opportunity and then customer case studies.

DOUG MILLER: So a lot of good information here, 22 case studies around customers who have migrated from UNIX to Windows, how they did it, why they did it, so lots of good information there.

Now Id like to bring up Marc Manley, VP of development at Rogue Wave Software, to talk about the new relationship weve just signed with Rogue Wave.

MARC MANLEY: Great, thank you, Doug.

As Doug just mentioned, I am VP of Research and Development at Rogue Wave Software. And I’m here to announce the agreement that we have with Microsoft, and to describe what this means to some of our mutual customers.

Rogue Wave is the global market leader in C++ software components. We’ve been around since 1989, and public since ’96, and our Source Pro C++ Technology is used extensively as the foundation for mission critical enterprise applications. We have a strong presence in financial services and especially on Wall Street, where our customers include all the major players. We’re also strong in telecom, aerospace, defense and software. Again, key customers are the large companies in all these segments.

Rogue Wave Source Pro C++ products give customers the flexibility to write applications once and then deploy those applications on the operating systems and databases of their choice. Source Pro extracts away the underlying details of these platforms, so customers can focus on implementing business logic, or in deployment where it makes the most sense for them.

For example, customers can write to our Source

Pro Database Abstraction once, and then just deploy the application on any of the major databases available on the market today. We also extract away the details of operating systems and hardware platforms. So, again, customers can write an application once using our Source

Pro product, and deploy that application on the operating system of choice, be that Linux, Windows, or a variety of UNIX platforms.

And in this way our Source Pro product makes it easy for customers to migrate their applications to different platforms. Today, there’s a large base of Rogue Wave-enabled applications running on UNIX servers, and many of these customers are evaluating moving to lower cost computing platforms, such as Windows on Intel hardware. And they need to preserve their investment in the current, in these economic times, as Bill was saying earlier, companies need to preserve that investment in their running applications.

Several of our mutual large customers actually brought us together because they were looking for greater flexibility to migrate applications to the Windows environment. The agreement between Rogue Wave and Microsoft that we’re announcing today is about improving the performance of our Source Pro C++ application on the Windows platform. This will give customers that same level of performance and portability on Windows.

Recently a team of Rogue Wave engineers met with Microsoft engineers to benchmark test the performance of Source Pro on Windows, and to define areas to improve performance. We’re implementing these enhancements now, and they will be available in our next version of Source Pro which will be coming out in October.

Our collaboration also includes joint sales and marketing activities, like today’s press conference, and we will be conducting joint meeting with our mutual customers and prospects.

Rogue Wave future technology direction is developing products that provide high levels of abstraction in the areas of .NET, Web services, and frameworks for developing large scale applications. We may be exploring other areas of collaboration with Microsoft. At Rogue Wave we’re very excited to be working closely with Microsoft, and looking forward to delivering greater value to our mutual customers.

As a result of this agreement, our customers will see stronger performance on the Windows platform, they’ll experience greater interoperability between UNIX and Windows, and they’ll have a viable option to migrate applications to the Windows Server platform.

If you would like more information, I’ll be in the back later on today. Thank you very much.

DOUG MILLER: Thanks, Marc.

We’re running a bit long in time, so I’m just going to take a couple of minutes to highlight two case studies. We have case studies at the back here, but two that are particularly interesting. One is an Italian insurance company, La Foundaria. These folks had 1,700 branch offices. They had a UNIX box and a Windows box in just about every office. They were looking to try to consolidate their systems, have only one system to worry about. The problem was, their UNIX system ran all their business operations, the Windows systems typically ran more of their office productivity applications. So, they wanted to build some new Windows applications but also leverage the existing UNIX apps. So, essentially they were able to bring over their UNIX apps, recompile them, bring them up on Windows, and were able to do that in about 30 days from initial proof of concept to fully deployed solution. The benefit there, of course, now they only have one server they need to deploy in each branch office, huge savings on administration and management, especially in such a distributed environment of 1,700 branch offices. So this was a great case study of somebody who was able to do this very, very quickly, and get huge returns on their investment very quickly.

Another one I would just like to quickly highlight, one of our server customers, the IRS.

BILL VEGHTE: Don’t hold it against us.

DOUG MILLER: Yes. They have been going through a modernization program for the last several years, and one of the things they wanted to do was get off of their UNIX desktops and get to a single Windows desktop. Most of them had desktops that were both UNIX, and then they had another machine that was Windows. And they only wanted to buy one machine for each agent. So, essentially, now they were able to move over their existing UNIX applications and run it directly on top of the Windows platform, and they’ve done something similar to what I showed earlier on, which is they extended that application, were able to wrap it, for example, with Word macros, and be able to call these old UNIX legacy apps through macros inside the Word applications. So, a very neat solution. Again, they were able to save a lot of money through this consolidation, and obviously save some time in administrating their desktops. So, again, the actual application port took three days, they were able to start rolling this out and field testing it in about three months.

So, I would like to hand back to Bill to wrap up, and we’ll be here for questions afterwards as well.

BILL VEGHTE: So, a couple of quick things from today. One is the momentum we have in the marketplace based on the scalability, reliability, and performance of the Windows platforms, and we see that, obviously, that trajectory continuing.

The long-term investment value that we afford customers and partners, that value is in business agility, in lower total cost of ownership, and in the opportunity that it affords partners and customers on competitive advantage.

And then we are a lower cost of ownership platform. PC economics, the ecosystem of partners that provide services to service and support that platform, we are a lower cost platform in these tough economic times.

And then, interoperability, interoperability is something that we have heard and understand is super, super important to you. Interoperability across a wide variety of clients, interoperability of data, interoperability of management, interoperability of network, interoperability around applications. Interoperability is super important, we understand that that’s something you need. Hence, our investment in SFU, and our partnerships with industry leaders such as Rogue Wave.

We’ll be up here for questions afterwards, I know we ran a little bit long. If you have any questions, the team will be up here to answer them. Thank you very much, enjoy the Krispy Kremes.


Related Posts