 Can we go to the big screen on the keynote or? Okay, here it is, here's Larry. I will never get tired of looking at that video. They actually, I got an email this morning and Judy Sim, our head of marketing, asked me if I, you know, which video I want to run. They found this other video that looked pretty exciting and they sent me a copy of it and I said, no, I think we'll stick with the one we ran on Sunday. It was a fantastic moment in my life. It was a fantastic moment in the life of the America's Cup and I am so enormously proud of the team that brought home the cup and some of the guys in the audience now and I would like them to stand up and just take a bow or just look back. Thank you very much for an incredible job. Go USA. Okay, this has been quite a week. If I could get the first slide please. We've announced more new technology, more innovation this week than any time before in Oracle's history. On Sunday night I introduced a new combination of hardware and software that we call the ExaLogic Elastic Cloud. And it's the second major product where we really have focused on integrating our software which we've been doing for years with the next generation of hardware technology and we believe fundamentally that if you engineer the pieces, the hardware pieces and the software pieces together, to work together, you get a much better overall system. By better I mean it's more reliable, it's easier to install and use, it's faster, it's more secure. Again, it's a little bit like the iPhone. I don't know if everyone knows it, but Steve Jobs is my best friend, I love him dearly. I watch very closely what he does at Red Apple. And he's believed for a very, very long time that if you engineer the hardware and the software to work together, the overall user experience, the product is better than if you just do a part of the solution. And actually most industries work that way. The car industry delivers complete cars. It includes the engine and the wheels and actually a lot of software. Whether it's navigation software or software that just computes how much gasoline is to be injected into the cylinders at that particular moment or what the temperature should be on your catalytic converter. There's an awful lot of computer systems running a modern car, but it all comes together. It's all engineered and tested to work together. Whether it's a Prius or, you know, my favorite commuting car at the Bugatti. So the exologic elastic cloud after the first machine that we introduced that combined hardware and software was Exadata, our database machine when we had the storage and the Infiniband network and the database servers and the memory systems and flash and all of those pieces combined with the Oracle database, our Linux operating system, an awful lot of software, an awful lot of software and hardware where all the pieces were engineered, tuned, fitted together and it delivered extreme performance and extreme reliability for our customers. And this week we've introduced kind of a new version of the Exadata machine, a new top of the line database machine using rather than two microprocessors per node, we have eight microprocessors or eight sockets per node. And that's the Exadata X2-8. So it's our new very top of the line Exadata machine, both the exologic machine and the Exadata machine, examples of what you can accomplish when you engineer hardware and software together and integrate all the pieces together. So the integration, a lot of the systems integration is done by us, the supplier rather than you, the customer, that should save you a lot of money and you should get a better and more reliable result. We'll be spending most of the time this afternoon in this presentation talking about Fusion applications. This is a project that began five years ago, more than five years ago. Actually, we knew we had to do this in December of 2004 when we completed the PeopleSoft acquisition. We actually got PeopleSoft and we got JD Edwards as a part of that acquisition. We suddenly had three different ERP systems, two different HRMS systems. We knew we were going to have to do a next generation and take the best ideas from PeopleSoft and the best ideas from the Oracle eBusiness suite and fuse them together in a next generation of technology. That was the beginning of the Fusion application project and today we're talking about the moments before we finally deliver Fusion to customers this year and make it generally available to all of our customers in the first quarter next year. So that's very exciting. It's the largest project, by the way. It's the largest engineering project in Oracle's history. We announced a new database, a new operating system kernel for our Linux operating system. As you know, about four years ago we went into the Linux business and we started by saying we're going to be 100% Red Hat compatible. We're not changing that, by the way. We still have the Red Hat compatible version of Oracle Linux, but we're introducing a new, highly optimized, high-performance version of our kernel, highly reliable version of our kernel called the Unbreakable Enterprise Kernel. And we had to do that, really, because we needed an operating system for exologic and exadata that was a lot faster than what we get if we stuck to strict Red Hat compatibility. We need an operating system that was much more secure and more reliable than we could have ever achieved if we had stuck to just Red Hat compatibility. So, again, from a Linux point of view, you want Red Hat compatibility, we got it. You want something that's faster and more reliable and more secure, we got that, too. We've introduced the next generation of Solaris, Solaris 11. Solaris has long been the most popular Unix in the world. It is the best enterprise-class Unix on the planet. We're very excited to now offer two different operating systems, Linux, which we've offered for four years, and now with the Sun team joining Oracle, we now have Solaris, the most popular, the highest quality, the most feature-rich, the most reliable, the fastest Unix on the planet Earth. Very excited about that. We're introducing, we've given you a roadmap, and we're introducing a whole new generation of Spark microprocessors. And our goal is to offer you more performance, more throughput through more cores at a lower, you know, in terms of TPMC per watt. You're going to hear me talking about TPMC per watt. How many transactions we can do while consuming the least amount of energy. And our goal with Spark, and we have a lot of them, is to be not only deliver the most throughput of any microprocessor on the planet, but do it while consuming the least amount of energy per amount of work done. And we've laid out the roadmap for Spark going forward. So those of you who are heavily invested in Spark and Solaris, there is going to be a next generation and a generation after that and a generation after that. Where the systems get more feature rich, faster, and more energy efficient. Java. We've given you a complete Java roadmap and a couple of things I'm most excited about are our new vector graphics. Graphics controlled by software. And we have 2D graphics coming out very soon and a little bit after that, 3D graphics. So much more capable, much more capable graphics capability built right into Java. So as you look at the next generation of UIs, which we think will be a combination of HTML5 and Java. Java is really designed to coexist with HTML5 and deliver a high quality graphical user experience from point of view of Java on devices. Obviously Java on servers is going to get faster and more reliable. Speaking of faster and more reliable, we introduced MySQL 5.5. Candidates 5.5 I guess is the right way to describe it where we've delivered huge performance improvements to MySQL. Huge reliability improvements to MySQL. All measurable stuff by the way. All absolutely measurable stuff. I know a bunch of people were concerned when we took over MySQL if we were going to kill it. And no, we want to make it better and make money. So rather than killing it, since we spent all that money to buy it, we'd like to get a return on that investment and the only way we can get a return on that investment is make MySQL better for you. Make it more appropriate for running your applications. So if it's delivering more value, then perhaps we can benefit by making MySQL a more profitable business unit for us. And this is just a partial list of all the things we've announced this past several days. This is dramatically more technology than Oracle has ever had. Our annual investment in R&D is now going to exceed $4 billion per annum and why stop there? So we're going to keep going, we're going to keep investing, we're going to keep innovating, and we're going to try to deliver more value than any other technology company. And our strategy is really quite simple. Our strategy is to take a lot of separate pieces that our customers used to buy as components and take those pieces and do pre-integration, the software pieces and the hardware pieces. Let's get those pieces integrated together and deliver you complete working systems like Exadata and Exologic. We think that will make your life simpler, our customers' life simpler, deliver you again something you plug in and work the first day. It's reliable, it's secure, it's fast, and it's cost-effective. Next slide. So we have better technology and more technology than ever before, and we have a more experienced management team than ever before. I'm excited that Mark Hurd has joined us from Hewlett-Packard, Joanne Olson has joined us from IBM, and Chuck Rosewatt has come back. He took a year off and went to Harvard, and we're thrilled that after that sabbatical and he's decided to come back and run support forces, everyone probably knows, Chuck ran all of engineering before Thomas Curry took over engineering, and now he's, again, after one year, I guess Chuck went to MIT, and I guess he was, when he was younger, and I guess he decided, gee, what was it like? Harvard was in the neighborhood. I guess he just missed out not going when he was younger, so he went this past time. I guess he only goes to school in Boston or Cambridge, whatever. But we're thrilled to have him back. He obviously knows Oracle very well. He's done a terrific job of building technology over many, many years, and during that time, I know he's worked with a lot of customers who had issues, and a lot of those issues eventually devolved and went into our development team. Chuck did a great job supporting those customers, and now he's doing that full-time. Again, we're thrilled to have Mark and Joanne and Chuck with us that we've never had a more experienced, more talented management team than we have today, and that's a pretty good combination. A lot of great technology, a lot of great people, and we think that that's a good recipe for us to deliver real value to our customers. Next slide. I'm going to be very brief, but I just want to repeat a little bit of my presentation on Sunday, which is the exologic cloud in a box. I have to chuckle a little bit about the phrase cloud in a box, because I know that the CEO of salesforce.com said, you know, Larry just doesn't get it, clouds don't run in a box. Okay? So what does he think salesforce.com runs on? If not on a box. Doesn't run on a box. Actually, salesforce.com runs on 1,500 Dell servers. Which are boxes? 1,500 of them. Now, he really got upset because we don't have an exologic box here on stage, but there are a lot of them in an exologic box here on stage, but the exologic box is about this tall. And he was really offended the box was taller than he was. And he said, clouds especially don't run on boxes that are tall. I mean, please look this up. You can't make this stuff up. I mean, okay. Do you think those 1,500 Dell boxes are all really like loaded to ground? You need acres and acres. If you come up, could you use less floor space unless you have really low ceilings? I mean, the thing that's interesting, let's look at the next slide about this. What the exologic machine is, it's 30 servers, Infiniband networking, a lot of storage, a virtual machine, operating system, middleware that runs all of your applications including salesforce.com. This is actually kind of the ideal box. If you wanted to go from 1,500 servers to 400 or whatever, I don't know what the right number is, but certainly a lot less than he's got, he would pick exologic. It's kind of the ideal machine to run an application like salesforce.com. But somehow I go back to the phrase Larry, you don't get it. A cloud is a box. This is the problem I've had with the term cloud computing. I think we made a big point of defining what we mean by cloud computing. We mean what Amazon means by cloud computing. We mean it's a platform that means hardware and software. That's right, a box and software. That's what it means. I'm sorry. It's a computer, it really is a computer. A cloud, by the way, folks, it's a software. It's a lot of boxes and a lot of software. We think the software has certain characteristics. We think it has to be virtualized. It has to be elastic. We made a bunch of stages of what we think cloud is. But our view and Amazon's view are pretty much the same. We kind of took what Amazon had, made an Oracle flavor to it, and put the hardware in the software so our customers can build private clouds out of this box or a company like Salesforce or NetSuite or Oracle Corporation can use these boxes for building public clouds. But I'm sure when Mark gets back and talks to his technical people they'll let him know, in fact, that you do need boxes. You really do. And what we're doing is we think the box should be efficient. It should be fast. It should be reliable. If you engineer the hardware and the software to work together in the box, you're going to get a much better experience. And I think the Salesforce customers and Salesforce.com would both benefit if they took a close look at Exologic and God Exologic and Exadata to run their cloud. And okay, I'll stop right there. Next slide. The Exologic cloud, it's 30 compute servers, 360 cores. All these things are interconnected via Infiniband. And we have an integrated software appliance that stores the software and application files. The interesting thing about this and I have a slide on it is when you patch when you patch the microcode, the virtual machine, the operating system, your applications you download one file to that storage appliance and you patch everything together. So not only is this thing easier to install, it's much easier to patch. It's much safer to pass. It's much more reliable. Again, I'm going to come back to this theme over and over again. You'll see banners. Hardware and software engineered to work together. We do the integration. We deliver the complete box. It saves you a lot of money and delivers much better performance than you otherwise could get. Next slide. The software we have in it is the WebLogic server, a very interesting piece of software for memory coherence. As I mentioned, we have 30 separate servers in that machine. And this coherence software kind of creates the illusion that those 30 separate servers is just one big memory system. It maintains memory coherence across the servers. JVM and Oracle VM Linux and Solaris. Next slide. It is incomparable. And this is why I think it's kind of ideal for Salesforce.com or NetSuite or us or anyone who, you know, it is so much faster than using conventional machines. So if they want to go from 1500 boxes to a few boxes, they really should look at this. It's elastic capacity on demand. This is really interesting. A modern cloud, if you build the cloud properly out of the right boxes when someone runs a report that really is consuming a lot of resource, you simply add more virtual machines to handle that additional capacity. What Salesforce has to do is have people canceling those reports if you get a report that's consuming too much because it's not a virtualized system. It's not a virtualized system. It doesn't have elastic capacity on demand. When demand goes up, their strategy is reduce demand, cancel reports. This thing has elastic capacity on demand. It's fall tolerant and scalable. Another thing Salesforce has that they need to upgrade is they have this thing called multi-tenancy. A lot of people say, oh multi-tenancy, that's what makes it cloud or that's what makes it software as a service, multi-tenancy. Multi-tenancy is a horrible idea. Multi-tenancy means every customer has all their data in the same database. One database, all the customers dump their data. GE, Siemens, you name it, all your data, all your data is commingled. Everyone's customer lists are in one single database. That is a horrible security model. In the 21st century, the way we support lots of different customers is called virtualization. Not multi-tenancy. Multi-tenancy is a decade old 15 year old technology. Virtualization gives you the ability to support lots and lots of customers in a secure fashion. Your data is separated and your own virtual machine, there's fault isolation, there's data protection. Multi-tenancy is old technology. If Salesforce upgraded to this, they could have a virtualized system unlike what they have now. And it would be fault tolerant unlike what they have now. It would be scalable and secure unlike what they have now. They need to look at something that's not 15 years old. Next slide. And if you want to run fast, this is the way to do it. That's the logic. In terms of HTTP requests, web requests, over a million, this one box, one 19 inch rack has over a million requests per second. Conventional hardware, that's 10 fold faster than conventional hardware. Ten times of performance improvement when running Java. In terms of messaging, almost two million messages per second. That's nearly a five times improvement over conventional hardware and conventional software. So this machine delivers enormous performance gains and cost savings because you need 20% as much hardware and software to do the same job. So it really is a huge cost savings, huge performance improvement, and of course consumes less energy. Next slide. It runs all your applications. So if you're not salesforce.com, you don't have a big application, but you're a conventional end user customer, you build this private cloud, and you can run everything on the exologic private cloud. The e-business suite, Siebel, PeopleSoft, your custom applications it all runs on the Exadata elastic cloud. It really is the right platform for delivering the benefits of hardware and software engineered to work together. And it's fully fault tolerant. It is fully fault tolerant. There is no single point of failure in either the Exadata or exologic boxes. If a server fails, it keeps running. If a network connection fails, it keeps running. If a virtual machine fails, it keeps running. It's fault tolerant not only the hardware failures, but also software failures. And we do that. That's what we engineer into the box. I've said it a hundred times, I'll say it again. If you engineer the hardware and the software to work together, the virtual machine to work with all the different servers, you can engineer these systems that are at once high performance, low cost, and they just keep running if there are failures. They tolerate failures. They not just disk failures, but CPU failures, memory failures, networking failures, complete server failures. Power supplies go out. This thing just keeps running. There are no single points of failure. It is an absolutely fault tolerant system built into the box. Next slide. You can start with a very small quarter rack and build up to a full rack of exologic boxes. Next slide. And it scales up to eight racks. Next slide. All the exadata and exologic boxes are identical. This is another important issue of hardware and software engineered to work together. Most of our customers have unique hardware software configurations. You make a decision on what networking cards you're going to have, what networking drivers you're going to run, what version of the operating system you're going to run, what VM you're going to run, virtual machine you're going to run, what version of the Oracle database you're going to run, what version of the Oracle middleware you're going to run. You make lots and lots of applications you're going to run. You make a lot of separate decisions. What version of this Dell CPU or this HP CPU or this IBM CPU you're going to be running. And then you put all these things together. Well no one's tested that configuration other than you. Our large customers all have unique hardware software configurations. They're all different and we've never tested that exact configuration. That's a problem. It doesn't mean it was not going to work. It just means when you deliver those pieces and you assemble those pieces and you integrate all those hardware and software pieces, you have to do a very thorough job of testing those components that have never worked together in exactly that way. Contrast that with the idea of taking letting us pick the VM and the version of the VM and us pick the OS and us pick the middleware and us the microprocessors and the memory systems and the flash cards and the disc controllers and the disk drives and there are so many different pieces and the networking cards and the networking fabric and the networking drivers and the disc drivers and the microcode the database I can go on and on letting us pick a single configuration and test it for millions of hours and make sure it works before you get it. That's the idea of hardware and software engineered to work together. It's that we can do all of this testing, all of this optimization all this performance tuning before you get it. The configuration is done. We know the pieces work together and work together well and then we deliver that same exact configuration to hundreds and thousands of our customers where your configuration is not unique anymore. Your configuration is identical to thousands of other configurations all over the world. Why is that good? It's not just that we can thoroughly test it. When a customer in that configuration has a bug on that configuration we know there are 30,000 other customers with that same configuration they will experience that same bug unless we notify them and prevent that bug from being rediscovered. It's not just that we do thorough testing it's just when another customer experiences a bug you're likely to experience the identical bug because you have the identical configuration. We can then come up with a patch and patch everybody's configuration safely to minimize bug rediscovery. About 98% of our support about 98% of our support calls are bug rediscovery that are bug related are bug rediscovery. We can cut way back on that by having hardware and software integrated to work together. Next slide. All the customers run the same configuration when we engineer the patch we don't just patch that one thing we run all of our regression tests for that configuration the exact configuration that you have so you can take that patch with confidence rather than getting that patch and say oh my god I now have to start thorough patch testing which all of you do you have to test this patch in your unique configuration because you know we couldn't have tested that patch in your unique configuration because we don't have a copy of your configuration you're the only one in the world that runs that configuration so hardware and software engineered to work together has huge benefit for our customers when you install the system it's thoroughly tested when you patch the system you patch everything together VM operating system middleware applications all those changes are tested and we know they work together millions of hours of testing rather than what you get now which is on your configuration no testing at all unless you do it yourself next slide but if you want hardware and software engineered to work together if you want security and performance and reliability you have to be willing to spend less the exologic machine is a fraction of the cost of other if you want to call it big iron I mean exologic is big iron without a big price you say well gee that's a million dollars that's a lot of money it replaces many millions of dollars of hardware and software so it actually is much lower cost solution much lower cost solution than a configuration delivering similar performance plus it's fault tolerant and the old IBM stuff is certainly IBM's big fast machines are not fault tolerant they have lots of their old SMP machines old architecture have been around a very long time it's vertical scale up rather than horizontal scale out it's not parallel ours is parallel we have no single points of failure they're loaded with single points of failure we're faster more reliable more secure more defensive not fault tolerant next slide we also announced Mark Hurd and John Fowler announced the exadata database machine the new top of the line database machine and same theme next slide it is hardware and software engineer to work together delivers the best data warehousing performance by far and the best OLTP performance by far first ever database machine that runs OLTP all the other database machines are data warehousing appliances this is a general purpose computing system that does all of your workload OLTP and data warehousing faster than on any other computer you can buy next slide over at softbank we replaced a 60 rack teradata machine with three racks I know Mark would be upset because there's three racks there's even more than one but less than 60 that would be Mark Benioff by the way so teradata was running on 60 racks running 60 racks at softbank we brought in three exadata racks and depending on the application those three exadata racks ran twice to eight times faster than the 60 teradata racks that was kind of shocking don't you think 5% we eliminated 95% of the racks and still ran on average five times faster with 5% of the hardware we ran five times faster that's why exadata and exologic are pretty good at building clouds Mark Benioff has to replace boxes with a few exadata it'll go well this is really a stunning example of what you can accomplish when you engineer hardware and software to work together next slide the new exadata machine it's the first exadata machine where we've actually not used two socket servers but instead for the first time used eight socket servers well going into all the technical detail it allows us to tackle very high end OLTP machines OLTP loads rather a single box has 128 cores you can then string eight of these boxes together with Infiniband it's an enormously it is by far by far the most powerful OLTP database machine ever built loaded with DRAM each 19 inch rack 50 terabytes of DRAM most OLTP databases will fit in main memory it's loaded with flash it's really a fast machine next slide we got a lot of data compression built into the oracle database and by compressing the data we can put 50 terabytes of your data your data warehouse into just flash memory so we can store a 50 terabyte data warehouse in flash and we have incredible query performance versus the other data warehousing appliance you can see the difference between teradata net teaser which now belongs to IBM I guess teradata IBM net teaser and exadata and by the way this comparison our throughput is five times net teaser is not including data compression if you throw in data compression that five times improvement goes to 25 times improvement over net teaser next slide in terms of data storage you can see a single rack of exadata that goes into four net teaser boxes and even more EMC boxes or teradata boxes next slide and it's much faster than any disk array it's much faster than EMC's disk array five ten times faster the thing about our storage system on exadata is we have a lot of parallel infinite band connections to the disk array and going into our database servers so we parallelize infinite band moving the data from our disk arrays into the storage servers so we move data 10 times faster 20 times faster 30 times faster then you can move data off an EMC disk array it will never be competitive because we engineered the storage servers in exadata and that's why we use this parallel infinite band fabric this is all plus we moved a lot of database software out of the database and into the storage servers which you don't get with EMC they can never build anything like that because we engineered re-engineered the database took it apart, moved some of it into the storage servers built a parallel interconnect between the storage servers and the database servers using infinite band and the bandwidth we get really is conservatively 50 times faster than what you get when you talk to an EMC disk array or a net apps disk array or anything like that it's a stunning difference and I'll come back to this theme it's what you can achieve if you engineer hardware and software to work together we encrypt everything now with exadata the encryption is all done in hardware hardware software working together is the theme so there is no risk you back up your database that whole thing is encrypted if those tapes disappear good luck in figuring out what's on them no one is going to be able to figure out what's on them because we have encryption built into the hardware there's no overhead anymore there's no overhead no performance penalty at all to encrypt all the data in your database why wouldn't you do it then when you back it up and someone misplaces one of those tapes and we read about that periodically in the newspaper who cares good luck figuring out what's on those tapes you're not going to lose credit card numbers or social security card numbers or health records you're not going to be able to do that stuff because your database is completely encrypted always everything built into the hardware IBM Netiza can't do that next slide and it's dramatically lower cost than other iron whether it's IBM's Netiza iron or IBM's power servers this costs a tiny fraction a tiny fraction of what it costs to do the same job using IBM iron and conventional storage so you save a lot of money you have a system that's more reliable if you use combinations of hardware and software that were carefully designed and engineered to work together next slide we have a complete family of exodators we've been in the exodator business now for a couple years and it starts with a quarter-rack of the two-socket database nodes and goes all the way up to eight racks of our new X2 Model 8 with the eight socket nodes so from a fairly low cost small exodata server you can grow that up to by far the biggest OLTP machine that has ever been built next slide the next slide we entered the Linux business about four years ago and we now have 5,000 customers it's gone very very well and our pledge to our Linux customers is that we will be 100% Red Hat compatible we promise so you can gracefully switch from Red Hat to Oracle Linux and all of your applications will run unchanged virtually nothing will change except we'll do a better job of fixing bugs that's it we'll do a better job of fixing bugs we'll do a better job of supporting you but all of your applications will run unchanged and will provide you with very very high quality support at a very low cost if you pick Oracle Linux and we convince 5,000 people to go that route and there is never once been a compatibility bug reported to Oracle that says hey you said that Oracle was compatible with Red Hat but here's an example where it's not that has never happened in four years of us supplying Red Hat compatible Linux to our 5,000 customers next slide but there are issues with Red Hat being compatible with Red Hat Red Hat happens to run a four year old kernel and I want to go into all the details Red Hat is really not designed for running something like Exadata or Exologic or high end fault tolerant computing high end secure fault tolerant computing with hardware accelerated encryption for example it's just not designed to do things like that so what we decided to do was to develop on Red Hat compatibility we're going to keep doing that but we're also going to do a second kernel so we have one version of Linux with two kernels you decide which one you're going to use when you boot up the system you can use the Red Hat compatible system Red Hat compatible kernel and get absolute Red Hat compatibility with Oracle fixing bugs and Oracle fixing and Oracle support or you can get the new kernel that's optimized for Oracle that delivers much more performance much more reliability and much better security and we now have these we now have these two kernels and next slide and the new unbreakable enterprise kernel is way faster way faster than Red Hat there's obviously a theme to what I'm talking about which is performance performance performance the faster we go the less hardware you need the less energy you consume the less you spend the less floor space you use up so it's not just trying to run your applications faster we like that too but we'd also like to make sure that you don't have to buy if you can buy 500 servers that's better than having to buy 2000 so over five times improvement in reading out a flash memory that's very important for Exadata and Exologic and it's going to be very important for other applications too is all applications are going in the direction of exploiting flash technology much better for SSDs SSDs two and a half times faster InfiniBand messaging three times faster and just OLTP performance almost twice as fast on a large server like the 8 socket machine and the Exadata X2-8 so whichever you choose the strictly red hat compatible kernel or the oracle enterprise kernel we give you a choice of strict compatibility or high performance reliability and security we highly recommend you use the unbreakable enterprise kernel for your oracle database your oracle middleware application next slide okay we've been working on building fusion applications for a very very long time more than five years next slide we have a lot of experience building ERP CRM and HRMS applications and we had a number of goals when we started the project to build the next generation of ERP CRM and human resources applications one thing we wanted to do by the way which had never been done before one thing we wanted to do was to use industry standard middleware to build these applications you know the e-business suite the oracle applications used to run on its own middleware which is ironic because oracle had a team of people building middleware that we sold to our customers and we had a completely separate team of people at oracle building middleware to support the e-business suite we had two separate middleware teams complete duplication of effort and products one aimed at our applications one aimed at our customers building custom applications at their shop at your place by the way peoplesoft was the same we bought peoplesoft and they had a huge middleware team building people tools they didn't use industry standardware or tools to build their applications siebel had something called siebel tools SAP has nobody nobody until now has ever been successful in building large scale ERP applications on top of industry standard middleware and we were determined to do whatever was necessary to make sure that we were going to deliver our applications that ran on industry standard Java middleware technology absolutely committed to that as a design principle the next thing the next goal we had we're back we're going to cover the keynote we're being kicked out of the floor here Dave but Larry Ellison is closing out his keynote all the good stuff out of the way but Larry Ellison just on fire throwing everybody under the bus EMC, NetApp Dell, IBM Tisa, Salesforce Red Hat Salesforce Mark Benioff basically came out today and said the cloud doesn't run in a box and Larry said well of course the cloud runs on a box and if you ran your cloud on our box you'd have a lot less boxes and those are Dell boxes big partner Michael Dell was here this week guess what Dell thrown under the bus he trashed Mark Benioff Benioff is getting bad advice I don't know who he's talking to but he's taking Larry Ellison head on he's got a ton of cash but he just killed everybody and he's coming out Larry's way is the highway and we saw the keynote let's just break it down America's Cup performance spraying champagne on everybody that's what he does you know and then ultimately he laid it down the Apple reference his friend Steve Jobs he called him my friend Steve Jobs he built hardware this is what they're doing it's an Apple generation Larry Ellison officially has Apple NV he's always admired Jobs we've talked about it the iPhone of the data center is what Exadata is and the Exologic Godbox it's all integrated hardware software database everything server storage flash to me the big story here is Larry Ellison has Apple NV he wants to be the Apple of the enterprise and the difference is Apple's a great product okay people love Apple do people love Oracle people love Exadata people love Exadata it's a small part of the portfolio now but I don't know about Exologic you know we'll see some of the other highlights he talked about Exologic was very proud of Exadata and then Exologic he talked about Fusion five years ago basically it's six by my count but people softened JD Edwards the new future fusing it all together on the hardware platform on Java which is the big reason why they bought Sun the largest engineering project in the history of Oracle available first quarter next year it's clear who runs engineering at Oracle hype dream or reality well I think you're going to see it eventually but I don't think you're going to see it there's no way you're going to see it this year I doubt next year I think reality 2012 he hands that he's going to go sailing, go mark her, I'll be back first quarter well I think it basically he'll keep people locked in in that time other highlights OS change around Exologic Solaris spark upgrades Java 3D capability MySQL they're addressing it new Linux kernel actually Larry said I know a lot of you were concerned that we were going to get rid of MySQL that we're going to kill MySQL he said no we want to make a profit out of MySQL Larry want to make a profit there's a shock big technology push from Larry around the 4B 4 billion in R&D the strategy he played out was we've been selling components for years we know what's the best sellers we're going to integrate them all together that was key and that's what he said we're going to bundle that that is the strategy, total bundling throwing everyone under the bus who's next? Accenture? that was the big message at this show and has been integrated hardware and software runs better, faster performance we have one minute we're going to wrap up here basically, Java on demand Java middleware multi-tenancy he said was a joke bad models about virtualization he was talking specifically about Salesforce there the key here is Apple Envy and that he is taking a hardware software integration approach and the other thing that's clear in this event Oracle Open World whole different vibe in terms of the ecosystem Larry's basically saying look you want to partner with me that's great love to have access to your customers and you know just the trashing of Mark Benioff just was amazing he's just like Mark Benioff and him are very competitive Mark used to work at Oracle Salesforce very aggressive he took him on, just threw him under the bus and he threw EMC under the bus EMC is one of the largest Oracle customers on the planet, Larry doesn't care he said we're going after you we're going after VMAX we're better, remains to be seen a lot of holes in Oracle's storage strategy the whole ecosystem is changing we talked about this early on day one this new strategy is all about changing the ecosystem a new configuration amongst the players this could be a very Cisco like move which we've been blogging about Cisco trying to take on the world Larry has got his eyes set on one thing world dominance, Apple Envy you know my take on this is the ecosystem is certainly rooting for and supporting VMware, very collaborative Oracle, it's like the mafia Oracle, Larry Ellison crushing it in his keynote he is so much fun to watch I love the American Cup he keeps us guys in business we're going to have food for fodder for him for months he always says a lot doesn't he live on SiliconANGLE thanks to QLogic, live on SiliconANGLE.tv we are at Oracle open world everyone's breaking down Dave, we're closing it out but we would not be able to bring you three days of continuous coverage if it was not for QLogic that's the proof for us thanks to JustinTV SiliconANGLE.com, JustinTV for putting this up in a reliable connection we have some bugs, we'll fix them out we'll talk to those guys, Dave thanks to you for coming in from Boston thanks for having the Wikibon community on here John, I appreciate it thanks to Markers and Hopkins Michael Sean Wright directing behind the scenes Kristen Nicole and Burt Latimore just filing stories all week. It's great stuff. The Cube will be there. We're looking at Hadoop World. I want to thank everyone out there who's watching for supporting us. We really appreciate it. Have a good night. You never know where the Cube is going to show up. Closing down from Oracle Open World, the Cube might show up in your area, and if you can get inside the Cube, bring your knowledge, and we will extract it and broadcast to the world. Come see us.