Slashdot Mirror
DOJ Gives Oracle Approval To Buy Sun
k33l0r writes "The BBC is reporting that the US Justice Department has approved Oracle's takeover of Sun Microsystems. The acquisition gives Oracle control over (or a leading role in), among other things, Java, MySQL, (Open)Solaris, ZFS, OpenOffice, and the NetBeans IDE. 'The European Commission has still to rule on the deal, a step that will be required before it can close. That body has indicated it will issue an initial opinion on Sept. 3, according to the Wall Street Journal. It may OK the deal at that time or launch a four-month probe of it. ... The Justice Department ruling came earlier than expected, a possible response to Sun's declining revenues and precarious business position in a steep recession, as the required reviews proceeded.' We first discussed the deal back when it was announced in April."
Apollo.
As far as mergers go this is probably a good fit. Oracle and Sun always needed each other for the most part. However I feel both are a dyeing breed. The industry wether you like it or not is moving away from those two companies core competencies.
High End Servers which are highly scalable with high end software which is highly scalable, is no longer the way it is now. We are moving to more smaller systems and don't need such scalability features as we realize that cost benefit really isn't there, for most situations.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
The only thing I'm concerned about regarding this deal is how this will change Java. The way I see it, one of two things will happen: One, current Oracle staff will manage the Java platform development and bad things will happen (all sorts of bad things could happen). Two, Oracle will deem Java an unprofitable product and will spin off a free software foundation, the likes of Mozilla or Apache.
For those wondering why the merger wasn't simply rubber stamped, it has to do with the licensing of Java:
http://www.jroller.com/scolebourne/entry/no_java_7_us_doj
From what I read, it wasn't a *huge* deal, but enough of a concern that the DoJ had to work with Oracle instead of simply approving the merger right away.
The EU probably has similar concerns.
This space left intentionally blank.
Or something like that.
A sad day.
---- Booth was a patriot ----
It's not quite as Larry put it in a corporate email he sent on the subject.
Well, I suspect that Oracle will attempt to position MySQL as their "free Oracle-compatible" database offering, less support and high-end features but still feature-complete enough that people will continue using it (and hopefully, in Oracle's eyes, upgrade to their full database suite when need arises).
/Mikael
Greylisting is to SMTP as NAT is to IPv4
Sun's x86 server hardware is quite competitive with Dell, in my experience.
http://rocknerd.co.uk
Unfortunately said Pentium 4s also would fail 10x more often.
I don't know if you've worked (ie, have had direct administrative experience) with any of the larger Sun hardware such as E2900 and above, or even the Ex500's from back in the day, but if you did you'd also know that these servers have a knack for uptime and resiliency that x86 servers, even to this day, have never had. There was a reason for those higher costs.
More to the point, I suspect they will position Oracle as a reliable MySQL-compatible database. MySQL uses a lot of weird extensions to SQL and owning the MySQL front end will make it easy for Oracle to add a 100% compatible front-end to their database. This will make it easy for companies that have deployed various things on MySQL to consolidate them all onto one big Oracle appliance (and, coincidentally, pay Oracle a lot of money in the process).
I am TheRaven on Soylent News
By "we" you must be referring to whatever market you are in. From my perspective, the amount of data being processed has increased and scalability is more necessary now than ever before. Large companies are increasingly involved in data mining and other large scale statistical analysis, and the need for computer systems that can perform those calculations in a timely is continuing to grow.
Palm trees and 8
Data Mining and Business Intelligence doesn't need huge powerhouses anymore. A low end server can easily handle the Millions of records Databases now. The Mid Range can handle Billions. What is left for Oracle and Sun are the Trillions of records DB. Which most sectors don't use. Also with advances in distributed computing we rarely need to go high end for the Trillions of records.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
I have a few concerns.
Oracle does not have a tradition of giving away much of it's software. Sun by contrast has a lot of open source or free as in beer software. I am worried that Oracle will either kill or start charging for Java, OpenOffice, Solaris, VirtualBox, MySQL and other products based on it's own business interests. It's only natural for it to do so. With this aquisition, Oracle is in a position of great power. It can kill or alter the course of all the products of both companies. Absolute power corrupts.
For example MySQL and PostgressSQL are the only 2 viable open source alternatives to an Oracle DB for many systems. (There are critical systems for which Oracle is absolutely needed, but the percentage that could be served well by an open source alternative is probably significant). It is definitely in Oracle's interest to kill or dillute MySQL.
These posts express my own personal views, not those of my employer
They're competing products in the same way a ford fiesta and a ford super duty truck are competing products.
Do you even lift?
These aren't the 'roids you're looking for.
um....
apparently you have never worked in a business that needs large amounts of related data that is generated by hundreds of systems across a geographic area.
It's a bad deal for both companies.
The acquisition of Sparc and Solaris further estranges Oracle from Microsoft... Most of Oracle's revenues come from windows-based products and the Solaris portfolio isn't likely to change that. Likewise, they now become a competitor in Java vs. Dot-net. It isn't smart to step up from mere competitor to antagonist without gaining a massive new strength, and that didn't happen here.
Then there's Java. Drains quite a bit of cash without making enough money and Oracle as a company has the wrong temperment to maintain and improve a programming language anyway. Start charging enough to make money on Java and Java dies. Nor does having Java particularly complement Oracle's product line.
And mysql is a mess too. Improving it drains sales from their flagship database product... but failing to improve it causes a fork which loses Oracle whatever value owning Mysql had for them. Bad mojo all around.
The Sun/IBM deal would have been much smarter. IBM has a huge market for the likes of Sparc and Solaris. Better yet, they have demonstrated the wherewithal to take code they own and insert it into Linux. There's lots of stuff in Solaris to like, IBM is already weighing heavily on the side of Linux-based products and services and a well supported Linux on Sparc could save Sparc from oblivion and maybe even return it to being a growing market.
Meanwhile, IBM's database product (db2) never escaped its tiny niche. MySQL would be a great complement to their portfolio, moving them squarely into the mainstream database business.
Lastly, IBM actually has a need for Java given the breadth of hardware and OS platforms they sell. Write once run everywhere would be a huge benefit to IBM. It strongly complements the rest of what they sell, even if they never make a nickle off of it directly. Sadly, IBM can't rely on Java when it's controlled by a company as boorish as Oracle. It has to remain a minor player in their portfolio.
Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion.
If you're somebody like eBay, you really really need scalabaility, as you're doing hundreds on non-idempotent transactions a second.
One of my much smaller customers needs 128 cores to reach a reasonable rate of speed committing sales transactions for a single line of business, so this isn't limited to very large companies or those with large data, just anyone with reasonable sales volumes.
--dave
davecb@spamcop.net
I wonder if this might lead to a dual-licensing for ZFS so it might be possible to use in linux.
c++;
Actually the advantage is a fast backplane, not the memory. You may remember they bought the rights to the Cray asynchronous (really packet-switch-like) backplane quite a number of years ago, and have been expanding on it since.
It's easy to build a fast chip if it never has to maintain cache-consistency with anything off-chip. If it has to stay sane, even with only 64*4*2 = 512 threads banging on the same memory range, it not only takes an expensive bus, but it's also memory-transaction-rate limited.
That's why you read about transactional memory in Linux Weekly News: we all need it, SPARC and Intel both.
--dave
davecb@spamcop.net
It will be very interesting to see how that pans out. I rather like Open Office - it's quirky and kind of ugly, but it does work and its drawing tools are great for business graphics. but its presentation tool (competitor for PowerPoint) sucks even worse than PowerPoint, and PowerPoint is at an advanced stage of suckitude. That said, I hope Ellison sees the promise in Open Office and really runs with it. If he could make OpenOffice presentation better than Keynote, word processing better than Word, and spreadsheet better than Excel, I would pay real money for that.
Shoes for Industry. Shoes for the Dead.
Exactly. The need for databases has not gone away, but the upper bound for what you can do with cheap, commodity hardware and the likes of Postgres or MySQL is now higher than most projects will ever need. Numerous popular web sites run on a handful of well-specified but basically off-the-shelf PCs. Almost any in-house business admin application can be run this way, too.
I'd go further than the parent post, though, because I suspect it's just as bad at the other end of the spectrum now. Unless you're working for something like a bank, a government social security department, or a massive commercial outfit like Amazon, you probably don't need the high-end capabilities of software like Oracle any more. However, if you really are playing in that league, it's probably cheaper to buy lots of commodity PC parts and build your own cloud than to use expensive, high-end server kit from the likes of Sun. Likewise, if you're Google or Amazon, you have the resources to develop bespoke software tools to match your needs anyway (and if you're not quite Google or Amazon yet, you can lease resources from those who are).
It's hard to see how either Oracle or Sun has much of a top-end target market left for its traditional products. It would be interesting if they went for an aggressive mid-range offering though, aiming at providing a complete hardware and software platform for mid-large businesses that are fed up with Microsoft but don't want to outsource everything to the cloud either. Post-merger, they'll have a credible office suite, more database expertise than anyone else, lots of supporting back-end/middleware tools, and a programming language and client-side software platform that were tailor-made for remote deployment.
If it turns out that the market likes the benefits of centralised admin and remote deployment, but wants to keep control in-house rather than trusting (and paying ongoing fees for) third party services, then an Oracle/Sun combo that invested its resources smartly over the next couple of years should be able to compete very strongly. They might even be able to build a credible long-term business model based on support, consulting and customisation, rather than relying on relatively few sales of expensive hardware and DB licences.
Thought for the day: if this goes through, I'll be glad I don't have shares in SAP.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
Unfortunately said Pentium 4s also would fail 10x more often.
I don't know if you've worked (ie, have had direct administrative experience) with any of the larger Sun hardware such as E2900 and above, or even the Ex500's from back in the day, but if you did you'd also know that these servers have a knack for uptime and resiliency that x86 servers, even to this day, have never had. There was a reason for those higher costs.
At the same time, the application landscape changed to prefer scalability that allowed servers to be down without impacting the whole system. A single machine no longer was so important.
And up until a few years ago, people still went with Sun when they had a single important node.
The choice became between more servers that will crash slightly more often with less overall impact to the application, and more servers that will crash slighly less often.
Larger Sun hardware was never amazingly reliable anyway. What it did have over Intel hardware was a greater chance of indentifying why it crashed. An Intel machine that crashes randomly is not unheard of. I've only had a few Sun machines in my life that required me to change more than a couple parts to stop a random crashing problem. (Except the E10K and Ex500 series, which were particularly bad, and the 420R, which had bad hardware by design).
I'd go further than the parent post, though, because I suspect it's just as bad at the other end of the spectrum now. Unless you're working for something like a bank, a government social security department, or a massive commercial outfit like Amazon, you probably don't need the high-end capabilities of software like Oracle any more.
A lot of Sun kit is highly parallel, so whereas you can buy one non-SPARC low-end system to handle things just fine, it may be better to purchase something like a T2000 or T5120 and use virtualization to split things up. For the same rack space and power usage you can run a bunch of hostnames with less overhead than something like VMware.
Of course Solaris runs just fine all all Tier 1 OEM systems (HP, IBM, Dell, Sun, etc.), so if you want to run x86 you can do that just fine and still get all the features of zones, DTrace, ZFS, etc.
CPU usage is not the only constraint you have to think about. Rack space, power, and cooling should be considered as well.
I just woke up and on reading the title I thought they meant Oracle was actually purchasing the sun...
How about changing Linux to use a less restrictive license?
"Data Mining and Business Intelligence doesn't need huge powerhouses anymore."
My theory about why has Sun Microsystems not done particularly well in the last few years is that the highly reliable hardware Sun Microsystems sells is no longer popular because it is far cheaper to use consumer-grade hardware with software that is fault-tolerant. The excellent 2008 book Planet Google describes Google's experiences on page 54: "For about $278,000 in 2003, [Google] could assemble a rack with 176 microprocessors, 176 gigabytes of memory, and 7 terabytes of disk space. This compared favorably to a $758,000 server sold by the manufacturer of a well-known brand, which had only eight multiprocessors, one-third the memory, and about the same amount of disk space."
But why would Oracle buy Sun? Possibly because there are difficulties in making Oracle database products work with the new consumer-grade hardware with fault-tolerant technology.
Changing the Linux license is legally impossible without removing a lot of the code.
c++;
Oracle does not have a tradition of giving away much of it's software. Sun by contrast has a lot of open source or free as in beer software. I am worried that Oracle will either kill or start charging for Java, OpenOffice, Solaris, VirtualBox, MySQL and other products based on it's own business interests. It's only natural for it to do so. With this aquisition, Oracle is in a position of great power. It can kill or alter the course of all the products of both companies. Absolute power corrupts.
Note: I do not work for Oracle, but we are a big customer of theirs. I have watched this very carefully, attended briefings (by Sun and by third party analysts.)
I am not concerned that Oracle will kill Java, OpenOffice, VirtualBox, MySQL. (I'm a little concerned about them selling off [Open]Solaris, since I don't see Oracle as an operating systems company.) However, I do expect to see a "pro" version of Java, OpenOffice, VirtualOffice, MySQL where Oracle forks the code into a stable branch, and companies can buy into a support contract for it. This isn't materially different from how OpenOffice/StarOffice are related now, or how Red Hat runs their business.
For example MySQL and PostgressSQL are the only 2 viable open source alternatives to an Oracle DB for many systems. (There are critical systems for which Oracle is absolutely needed, but the percentage that could be served well by an open source alternative is probably significant). It is definitely in Oracle's interest to kill or dillute MySQL.
I disagree that Oracle wants to kill or dilute MySQL. Quite the opposite, really. Oracle desperately wants to compete with SQL Server at the lower-end databases. Small companies and many mid-size companies feel that Oracle is much too complicated for them to run with [typically] a limited IT staff. Oracle has a lot of buttons, knobs, switches to tune performance (not to mention get things running.) As a result, SQL Server often gets deployed here. And for most internal-office workloads for small or mid-size companies, SQL Server works very well. So Oracle doesn't make money here. Oracle knows that lots of people can (and do) easily deploy MySQL, this is an easy "win" for them.
My $0.02
The head of Oracle and 3rd richest man in the world visited the lowly Java developers conference last month and gave full support for Java inside the new Oracle.
As you say, you can send every transaction to four machines, but that's not the whole story. First, lots of transactions need to be ordered. You need to make sure that query one from client one happens before query two from client two on all four machines. There are some quite simple algorithms that you can use for this, but they generally increase bandwidth usage and latency over just having a single machine. It doesn't take long for the cost of the software and the additional infrastructure you need to dwarf the savings you make by not buying reliable hardware in the first place. Software developers are expensive, and software developers that can write fault-tolerant code that you'd trust your business on are very expensive. A reasonable Sun machine costs a lot less than employing one of these people for a year.
I am TheRaven on Soylent News
Think you are missing one important point, yes the world is moving to smaller systems. But if you have a look at a four socket system with 8 core CPUs in there and add hyperthreading, you end up with a system that to the OS and the application looks like a 64way smp with a bit of numa. And lets see, that looks very much like a "large scalable" system, say 5 years a go. So the hw is getting smaller and cheaper, which allows us to build cheep "large scalable" systems, but as nice as it sounds now you will need an OS that can handle all of those CPUs, and lets see Oracle gets a OS that's been thriving on those types of systems for the last 10+ years, namely Solaris/OpenSolaris. I do believe that Sun would have done much better if they would have stuck with what they do best, build HW and write infrastructure SW to make that HW shine, rather than trying to become a SW company, of which they have shown a number of times they don't have a clue.... That is selling SW, they do write very good SW