Maybe this isn't out the of realm of conceivability to others, but it was to me...Oracle is a software company (one that runs a lot on Sun hardware), and suddenly becoming a hardware company has got to be a daunting challenge, regardless of who you are or how smart you are.
The implications are staggering across the board. Maybe Oracle decides they don't want to the hardware, just Java and MySQL (...they got it, finally), but then all that Sun hardware and Solaris...? Or maybe they want to make Solaris/Sun hardware the best platform for Oracle products (already the case as far as I know), then what of support for all their other platforms.
Oracle likes to buy a lot of companies, but they've all been, more or less, niche players in specific markets to fill in the gaps of their own offerings. I can't imagine what "gap" buying Sun will fill, other than something will be certainly be filled.
Oracle likes to buy a lot of companies, but they've all been, more or less, niche players in specific markets to fill in the gaps of their own offerings. I can't imagine what "gap" buying Sun will fill, other than something will be certainly be filled.
Application server? Java development environment? Control of the Java language? UI Technology? Hardware?
Everyone seems to be missing the big picture: Oracle's goal is to offer you a fully supported "stack" from database to application server to hardware and everything in between. All the development tools, technologies, languages, etc. So they can lock you in and offer you the full range of support, no handing you off to so and so because it's not a database problem anymore. Would you pay a premium for that? That's how you make money. And now, they have filled a lot of those gaps and have absorbed some great teams to make that dream a reality. Or so they believe. We'll see how this turns out.
If you have the money. For anything you would want to run in production it costs tens of thousands of dollars per server (assuming a quad core or dual quad core intel server). High tech start ups like where I am working now can't afford that kind of money, even with deep pockets backing it up. The other alternative is JBoss, which not everyone agrees is all that good, even if it does provide a good number of features. Also, since it is being developed by Sun, it can or will serve as a
You are assuming that from a customer perspective, a single vendor experience is superior to a multi vendor experience.
It's not so much that the commenter made that assumption but that the Captains/Clowns of Industry made that assumption. Those of us who live in the real world know two things: it depends on the vendor; and the vendor will acquire monopolistic tendencies in a hurry.
In the 'you scratch my back I scratch yours' world of extra big business, this isn't really a problem. They each have the same ambition to be a monopolist, so helping along a corporation (and its board of directors) that is making progress in t
Its highly unlikely Oracle will maintain Sun's hardware aspect of the business. Sun already has put SPARC into legacy mode. Oracle will probably keep or sell off the hardware products that can sustain itself. It will probably maintain the legacy server stuff, to keep its high-end ticket customers who buy Sun for high-availability systems.
An accepted tactic to grow a customer base is to buyout another company's customer base. Its usually considered to be a cheaper route than investing in taking away a competitor's customer base. This is probably the reason Oracle went for Sun. Oracle has become more services/consultant oriented. It can't really break into IBM's territory, partly because of IBM's hardware components for "complete solutions" or enterprise market. This allows Oracle to grab all the customers IBM hasn't already taken away.
The bigger question is what Oracle plans for Sun's software products, like Solaris, MYSQL, and Java.
I agree that Oracle will kill off Sparc, but I would expect it to retain a Sun-branded hardware business, based on Intel. That will be a key part of the soup-to-nuts stack strategy, I would have thought.
No, what would Oracle gain from selling customized PCs? (I exaggerate, but that's what it is.) It costs money to invest in hardware development and marketing to get companies to buy the hardware package over a cheaper competitor. Companies didn't buy Sun equipment for its hardware. They bought it as a component of an enterprise high-availability system. Sun didn't sell a database product. They sold a platform product to run your database product on. Sun depended on OS lock in to move their hardware pr
Sun's multicore sparc work is basically custom designed to run giant database servers, and giant web servers with giant database back ends. Doing so at lower power draw than the competition has the potential to be a market winner. That alone will not be sufficient, however.
That was a good theory for IBM's attempt to purchase Sun. It doesn't work so well for Oracle, since there isn't that much overlap between Oracle and Sun's product portfolio. Worse for this theory is that there is already a huge overlap between Sun and Oracle *customers*. So, according to your theory, Oracle bought Sun, so that it can sell to Sun customers products they already own. Try again.
Its highly unlikely Oracle will maintain Sun's hardware aspect of the business.
Dude, most of Sun's revenue comes from hardware, either by selling it, or by selling services bundled with it. Software people see sun in terms of Java, MySQL, and so on, but these actually generate very little revenue on their own; to Sun, they're valuable mainly as a way of driving its hardware and service businesses.
If Oracle writes off Sun's hardware business, they will have paid $5 billion ($7 billion plus the cost of assuming Sun's debts, but minus Sun's $3 billion
Safra Katz (Oracle President under Ellison) is saying that he can make Sun's hardware business profitable. That's a credible claim.
I'm surprised more companies weren't interested in buying Sun. They have over $13 billion dollars in revenue. A little bit of fine tuning in the business could mean huge profits.
Sun is phasing out sparc, but aggressively selling UltraSparc.
The cost vs performance is very impressive, and is winning people back from x86. At least in my environment (education).
The oracle database runs really well on the ultrasparcs. We just bought a 4 cpu T2 ultrasparc server, it has 8 cores per cpu for a total of 256 threads. With an edu discount, it cost ~60,000 if I recall correctly. There is nothing remotely as powerful in the x86 realm. http://www.sun.com/processors/UltraSPARC-T2/specs.xml [sun.com]
You're right. I just don't get it. MySQL makes sense for Oracle. The hardware business makes no sense whatsoever.
Java really doesn't make much sense for Oracle, either. A lot of databases might get use Java front-ends, but so what? Oracle hasn't been in that business.
In the end, I think goes down like this: This is about two things: Red Hat and MySQL. Oracle's RHEL variant has been a a complete bust; Oracle customers have been sticking with Red Hat. Read Matt Assay's column over on C|Net if you don'
You're looking at it all wrong. Oracle sells a database/services/consulting product, not an OS product. They would NEVER niche themselves into a complete black box product. They'll keep offering current products on Linux and on IBM. Oracle doesn't like RedHat because they think they can supplant RedHat's service offerings. The market is not going to move back to Solaris.
Java really doesn't make much sense for Oracle, either. A lot of databases might get use Java front-ends, but so what? Oracle hasn't been in that business.
Not so - Oracle isn't just a database company. Much of the rest of their technology stack relies on Java. Do a search for "Oracle Fusion".
They acquired a leading Java EE vendor in BEA a couple of years back.
Java really doesn't make much sense for Oracle, either. A lot of databases might get use Java front-ends, but so what? Oracle hasn't been in that business.
You're joking, right? Oracle owns BEA now, and therefore both their Weblogic app server and the JRockit VM. Oracle has the TopLink implementation of the JPA-compliant ORM layer. They have the JDeveloper IDE. Those are just the things off the top of my head without searching.
A Sun/Apple merger would have made sense ten years ago, when Apple had a great desktop UNIX but with an ageing kernel and running on CPUs from a company that couldn't meet their demands. Sun had a decent server UNIX, with a nice kernel, but no real presence on the (corporate) desktop. OS X on a Solaris kernel, on SPARC would have been very nice, and could have scaled right down to the SPARC v8 systems designed for handheld systems up to the v9 cores designed for massive SMP servers. Steve Jobs still hasn't forgiven Sun for abandoning OpenStep though, so it was never very likely.
The real shame is that, in the mid '90s, Sun put together an incredible hardware and software stack for mobile devices. A few bits of it made it into Java, but most of it never went to market. If Sun had licensed the software and sold the hardware to ODMs then they would almost certainly not have been looking for a buyer now.
I don't think OS9 was a UNIX-system. Ten years ago, that was Apples operating system.
Mac OS X Server 1.0 was released on March 16, 1999. Naturally, it wasn't a desktop OS; however, it was still NextStep (and therefore UNIX) based, and Apple did have it.
Why do you find more likely a merger of a desktop company with a server company, than the merger of two complementary server companies.
For the foreseeable future, nobody is going to pry the corporate desktop market away from Microsoft.
But in the corporate data center, merging the most popular enterprise database (Oracle), the most mature, advanced and popular Unix OS (Solaris/SPARC) and matched hardware, along with Java app servers and middleware to go head to head with IBM Global Services is going to be very profitable.
Not really that daunting. They can shutdown their hardware part and turn it into an intellectual property portfolio by throwing a bunch of lawyers around it. Then, they just target x86 while cross-licensing stuff. It'll take a few years to pan out, but it's a likely scenario.
"You can have my Unix system when you pry it from my cold, dead fingers."
-- Cal Keegan
Wow. Just Wow. (Score:5, Insightful)
Maybe this isn't out the of realm of conceivability to others, but it was to me...Oracle is a software company (one that runs a lot on Sun hardware), and suddenly becoming a hardware company has got to be a daunting challenge, regardless of who you are or how smart you are.
The implications are staggering across the board. Maybe Oracle decides they don't want to the hardware, just Java and MySQL (...they got it, finally), but then all that Sun hardware and Solaris...? Or maybe they want to make Solaris/Sun hardware the best platform for Oracle products (already the case as far as I know), then what of support for all their other platforms.
Oracle likes to buy a lot of companies, but they've all been, more or less, niche players in specific markets to fill in the gaps of their own offerings. I can't imagine what "gap" buying Sun will fill, other than something will be certainly be filled.
Re:Wow. Just Wow. (Score:5, Insightful)
Oracle likes to buy a lot of companies, but they've all been, more or less, niche players in specific markets to fill in the gaps of their own offerings. I can't imagine what "gap" buying Sun will fill, other than something will be certainly be filled.
Application server? Java development environment? Control of the Java language? UI Technology? Hardware?
Everyone seems to be missing the big picture: Oracle's goal is to offer you a fully supported "stack" from database to application server to hardware and everything in between. All the development tools, technologies, languages, etc. So they can lock you in and offer you the full range of support, no handing you off to so and so because it's not a database problem anymore. Would you pay a premium for that? That's how you make money. And now, they have filled a lot of those gaps and have absorbed some great teams to make that dream a reality. Or so they believe. We'll see how this turns out.
Re: (Score:2)
If you have the money. For anything you would want to run in production it costs tens of thousands of dollars per server (assuming a quad core or dual quad core intel server). High tech start ups like where I am working now can't afford that kind of money, even with deep pockets backing it up. The other alternative is JBoss, which not everyone agrees is all that good, even if it does provide a good number of features. Also, since it is being developed by Sun, it can or will serve as a
Re:Wow. Just Wow. (Score:5, Funny)
It's IBM without... you know... the "IBM."
Re: (Score:2)
really, the 'sell you everything in one stack' sounds more Microsoft than anything else.
At least Oracle's new hardware business will do better than Microsoft Bob and the Xbox.
Re: (Score:2)
Re: (Score:2)
You are assuming that from a customer perspective, a single vendor experience is superior to a multi vendor experience.
It's not so much that the commenter made that assumption but that the Captains/Clowns of Industry made that assumption. Those of us who live in the real world know two things: it depends on the vendor; and the vendor will acquire monopolistic tendencies in a hurry.
In the 'you scratch my back I scratch yours' world of extra big business, this isn't really a problem. They each have the same ambition to be a monopolist, so helping along a corporation (and its board of directors) that is making progress in t
Re:Wow. Just Wow. (Score:5, Informative)
Its highly unlikely Oracle will maintain Sun's hardware aspect of the business. Sun already has put SPARC into legacy mode. Oracle will probably keep or sell off the hardware products that can sustain itself. It will probably maintain the legacy server stuff, to keep its high-end ticket customers who buy Sun for high-availability systems.
An accepted tactic to grow a customer base is to buyout another company's customer base. Its usually considered to be a cheaper route than investing in taking away a competitor's customer base. This is probably the reason Oracle went for Sun. Oracle has become more services/consultant oriented. It can't really break into IBM's territory, partly because of IBM's hardware components for "complete solutions" or enterprise market. This allows Oracle to grab all the customers IBM hasn't already taken away.
The bigger question is what Oracle plans for Sun's software products, like Solaris, MYSQL, and Java.
Re: (Score:3, Interesting)
I agree that Oracle will kill off Sparc, but I would expect it to retain a Sun-branded hardware business, based on Intel. That will be a key part of the soup-to-nuts stack strategy, I would have thought.
Re: (Score:3, Interesting)
No, what would Oracle gain from selling customized PCs? (I exaggerate, but that's what it is.) It costs money to invest in hardware development and marketing to get companies to buy the hardware package over a cheaper competitor. Companies didn't buy Sun equipment for its hardware. They bought it as a component of an enterprise high-availability system. Sun didn't sell a database product. They sold a platform product to run your database product on. Sun depended on OS lock in to move their hardware pr
Re: (Score:2)
Yes. Kinda like IBM does.
Sparc into legacy mode? (Score:5, Interesting)
Didn't get that impression last time I attended one of their seminars a few weeks ago.
The multicore stuff Sun is doing is miles ahead og anything anybody else is doing,. I hope Oracle do not axe that.
sparc and oracle (Score:4, Interesting)
Re: (Score:2)
Sun already has put SPARC into legacy mode.
RedHat decided to switch to OpenSolaris.
See, I can make up stupid statements without anything to back it up too.
buying out the customer base (Score:2)
Re: (Score:2)
Hoo boy. Where do you get your facts?
Its highly unlikely Oracle will maintain Sun's hardware aspect of the business.
Dude, most of Sun's revenue comes from hardware, either by selling it, or by selling services bundled with it. Software people see sun in terms of Java, MySQL, and so on, but these actually generate very little revenue on their own; to Sun, they're valuable mainly as a way of driving its hardware and service businesses.
If Oracle writes off Sun's hardware business, they will have paid $5 billion ($7 billion plus the cost of assuming Sun's debts, but minus Sun's $3 billion
Re: (Score:2)
Safra Katz (Oracle President under Ellison) is saying that he can make Sun's hardware business profitable. That's a credible claim.
I'm surprised more companies weren't interested in buying Sun. They have over $13 billion dollars in revenue. A little bit of fine tuning in the business could mean huge profits.
Re: (Score:2)
Sun is phasing out sparc, but aggressively selling UltraSparc.
The cost vs performance is very impressive, and is winning people back from x86. At least in my environment (education).
The oracle database runs really well on the ultrasparcs. We just bought a 4 cpu T2 ultrasparc server, it has 8 cores per cpu for a total of 256 threads. With an edu discount, it cost ~60,000 if I recall correctly. There is nothing remotely as powerful in the x86 realm.
http://www.sun.com/processors/UltraSPARC-T2/specs.xml [sun.com]
Re: (Score:2, Interesting)
You're right. I just don't get it. MySQL makes sense for Oracle. The hardware business makes no sense whatsoever.
Java really doesn't make much sense for Oracle, either. A lot of databases might get use Java front-ends, but so what? Oracle hasn't been in that business.
In the end, I think goes down like this: This is about two things: Red Hat and MySQL. Oracle's RHEL variant has been a a complete bust; Oracle customers have been sticking with Red Hat. Read Matt Assay's column over on C|Net if you don'
Re: (Score:2)
You're looking at it all wrong. Oracle sells a database/services/consulting product, not an OS product. They would NEVER niche themselves into a complete black box product. They'll keep offering current products on Linux and on IBM. Oracle doesn't like RedHat because they think they can supplant RedHat's service offerings. The market is not going to move back to Solaris.
Re: (Score:2)
They would NEVER niche themselves into a complete black box product.
Uhm... That's exactly what they said they WOULD do in the conference call.
The market is not going to move back to Solaris.
According to Oracle, the market never moved away from Solaris/SPARC, it's still beats Linux for Oracle deployements.
Where do you get your information from?
Re: (Score:2)
Not so - Oracle isn't just a database company. Much of the rest of their technology stack relies on Java. Do a search for "Oracle Fusion".
They acquired a leading Java EE vendor in BEA a couple of years back.
Re: (Score:3, Informative)
Java really doesn't make much sense for Oracle, either. A lot of databases might get use Java front-ends, but so what? Oracle hasn't been in that business.
You're joking, right? Oracle owns BEA now, and therefore both their Weblogic app server and the JRockit VM. Oracle has the TopLink implementation of the JPA-compliant ORM layer. They have the JDeveloper IDE. Those are just the things off the top of my head without searching.
Re: (Score:3, Funny)
I think Oracle on AIX is pretty powerful, and popular.
Re: (Score:1)
I think Oracle on AIX is pretty powerful, and popular.
Shoulda stopped after "powerful".
Re: (Score:2)
as a user of oracle on AIX, I must agree.
Re:Wow. Just Wow. (Score:4, Interesting)
Re:Wow. Just Wow. (Score:5, Interesting)
A Sun/Apple merger would have made sense ten years ago, when Apple had a great desktop UNIX but with an ageing kernel and running on CPUs from a company that couldn't meet their demands. Sun had a decent server UNIX, with a nice kernel, but no real presence on the (corporate) desktop. OS X on a Solaris kernel, on SPARC would have been very nice, and could have scaled right down to the SPARC v8 systems designed for handheld systems up to the v9 cores designed for massive SMP servers. Steve Jobs still hasn't forgiven Sun for abandoning OpenStep though, so it was never very likely.
The real shame is that, in the mid '90s, Sun put together an incredible hardware and software stack for mobile devices. A few bits of it made it into Java, but most of it never went to market. If Sun had licensed the software and sold the hardware to ODMs then they would almost certainly not have been looking for a buyer now.
Re: (Score:1)
Re: (Score:2)
Mac OS X Server 1.0 was released on March 16, 1999. Naturally, it wasn't a desktop OS; however, it was still NextStep (and therefore UNIX) based, and Apple did have it.
Re: (Score:2)
You have to go back more than 10 years.
In the late 80s/early 90s Apple had some corporate marketshare, but by 1999 that was long gone and Apple was in fruity iMac mode.
Re: (Score:1)
Re: (Score:2)
Why do you find more likely a merger of a desktop company with a server company, than the merger of two complementary server companies.
Re: (Score:2)
Why do you find more likely a merger of a desktop company with a server company, than the merger of two complementary server companies.
For the foreseeable future, nobody is going to pry the corporate desktop market away from Microsoft.
But in the corporate data center, merging the most popular enterprise database (Oracle), the most mature, advanced and popular Unix OS (Solaris/SPARC) and matched hardware, along with Java app servers and middleware to go head to head with IBM Global Services is going to be very profitable.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)