BT

Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ

Topics

Choose your language

InfoQ Homepage News Perspectives on the Conclusion of the Oracle - Sun Acquisition

Perspectives on the Conclusion of the Oracle - Sun Acquisition

This item in japanese

Bookmarks

After almost nine months of speculation and delay, Oracle has got the green light from EU which has lead to the completion of Sun’s acquisition. The announcement was followed by an all-day event were Oracle presented its future plans for the Sun technologies and platforms.

Java, the JVM and Alternative JVM Languages

As Thomas Kurian, Oracle's senior VP of product development stated, his company plans to integrate the Sun HotSpot and Oracle JRockit Java virtual machines. He also said that Oracle intends to "revitalize" the Java developer community and extend the reach of the Java programming model to support emerging application development paradigms. For example Oracle plans to add modularity and support for multicore processing to Java SE and to provide new levels of support for Java ME with new features such as multitouch.

InfoQ has contacted Allex Miller regarding Oracle announcements for the JVM:

 It sounds to me like they want to take some of the cool parts of BEA's LiquidVM "JVM on a hypervisor" technology and merge that into the existing Hotspot code.  I think the virtualization bits are certainly some of the most interesting (and lucrative) parts of the JRockit JVM and play really went into the current trends toward virtualization, clouds, and clusters as a way of managing your compute resources.  This could be a significant competitive advantage against IBM's J9 JVM.  

I think the comments about removing permgen and using thread-local GC sound like great ideas as well.  The thread-local GC probably refers to the escape analysis and stack allocation optimizations that Hotspot has been working on for a while.  Most programs create many temporary objects that are used within the context of a single thread but never escape to the visibility of other threads.  Those objects can be allocated directly on the stack (which is faster) and do not enter the heap or require inspection or removal by the normal garbage collector (which reduces GC).

The permgen improvement is really needed to address a problem that has been getting worse with dynamic languages running on the JVM.  Many of these languages like Groovy or JRuby dynamically generate large numbers of small classes at execution time to provide the dynamic behavior.  These classes pollute the special "permgen" parts of Java memory and are difficult or impossible to reclaim, causing significant memory problems.  The JSR 292 invokedynamic work is addressing some of this problem in another way by allowing dynamic languages to directly link call sites at runtime, avoiding the need to generate many of those internal classes at all.

In general, I think it's great that they are not just dropping either of these JVMs but trying to take the best of both.  Both engineering teams have phenomenal people that have been doing innovative work for a long time and I'm glad to see that they want to continue investing in this area.  That should give some hope that the JVM will remain a vibrant and exciting place to develop cutting edge new languages like Scala, Clojure, Groovy, and JRuby.

Oracle has disclosed very little information on the future of the JCP and how it will go about developing Java 7, as Stephen O'Grady from RedMonk points out:

I think Oracle may be slightly more pragmatic about the JCP in general than was Sun, for whom the platform involved significant emotion, but it’s difficult to predict how things will play out.

Mysql

Edward Screven, Oracle's chief open-source architect, stated that his company will continue to support MySQL database which Oracle views as complementary to its core database technologies and not a competitive product. Oracle CEO Larry Ellison insisted that his company will do a better job of improving MySQL than its previous stewards, without mentioning Sun or the open source community. Oracle will offer MySQL through a separate sales team, while enhancing its compatibility with other Oracle software applications.

JavaFX and Rich Internet Technologies

During the announcement, Oracle has reassured once again that it will be investing heavily in JavaFX and will make significant investment towards the fusion of DHTML, JavaScript, Java, JavaFX.

This seems to fills a vacuum created when Oracle terminated BEA's former arrangement to bundle Adobe Flash/Flex development tooling, as Tony Baer from ZDNet points out:

We’re also not surprised as tot he prominent role of JavaFX in Oracle’s RIA plans; it fills a vacumm created when Oracle tgerminated BEA’s former arrangement to bundle Adobe Flash/Flex development tooling. Inactualityy, Oracle has become RIA agnostic, as ADF could support any of the frameworks for client display, but JavaFX provides a technology that Oracle can call its own.

Author and JavaFX advocate Jim Weaver feels confident about Oracle supporting the JavFX platform:

I am even more convinced by today's events that JavaFX will continue to move forward and be chosen with increasing frequency as the rich internet application (RIA) platform for new applications.  There are at least three big challenges for JavaFX that need to be overcome, so I trust that Oracle's resources will be focused quickly on these challenges:[…]

Netbeans

As InfoQ has reported in the past, the future of Netbeans is uncertain after the acquisition.

Tony Baer is certain that Netbeans will become a second rate citizen in favor of JDeveloper:

You can play around with NetBeans, which Oracle’s middleware chief Thomas Kurian characterized as a “lightweight development environment,” but again, if you really want to develop enterprise-ready apps for the Oracle platform, you will still use JDeveloper, which of course is written for Oracle’s umbrella ADF framework that underlies its database, middleware, and applications offerings. That’s identical to Oracle’s existing posture with the old (mostly) BEA portfolio of Eclipse developer tools. Actually, the only thing that surprised us was that Oracle didn’t simply take NetBeans and set it free – as in donating it to Apache or some more obscure open source body.

Similarly Stephen O'Grady believes that there is a significant chance that Oracle will not be investing in Sun’s IDE:

[…] Likely candidates like NetBeans or OpenOffice.org were explicitly mentioned on yesterday’s call, which presumably wouldn’t be the case if the plan was to immediately retire them. […]Yes [it will survive], although it sounds as if it will take a backseat to JDeveloper.

GlassFish

As Thomas Kurian, Oracle's senior VP of product development stated, Sun's Web app server will continue to get support from Oracle, but it will be offered primarily as a departmental solution, while Oracle's own WebLogic Server will continue to be marketed as the enterprise solution.

Stephen O'Grady feels that GlassFish will not be endorsed by Oracle:

It will be, as Oracle phrased it, a reference implementation. Beyond that, Oracle didn’t have much in the way of promises. GlassFish, as someone reminded me this morning, is not getting its own sales and marketing organization, as MySQL is. Which you can read two ways. First, as a commenter wrote, that “Oracle sees more opportunity for joint selling of GF+WLS to address the differing needs of projects.” Or, more cynically, that they see the product as potentially disruptive to their WebServer product line, and will kill it softly through organizational dynamics. How incented do you think the WebLogic sales folks are going to be to push a lower priced alternative?

The Cloud

Edward Screven, Oracle's chief corporate architect, has stated that this company will not be offering Sun's long-planned Cloud service. Sun had original announced its plans for an Amazon-style cloud with compute and storage that would support Linux, Windows, and Solaris on a mix of Sparc and x64 machines, using Sparc blades and both Xeon and Opteron processors on x64 blades as well as open-source products ZFS and Crossbow.

Sun's Cloud initiative had followed it older plans for a grid utility computing work, Network.com, which failed to attract customers and was eventually phased out in anticipation of the cloud.

Stephen O'Grady has commented on Oracle’s opposition to Sun’s Cloud vision:

Customers have overwhelmingly avoided appliances, both virtual and otherwise, in favor of so-called best of breed architectures. Still, as much as Ellison hates the cloud, it may help this time around.[…] Ellison hates the cloud because, in his view, it’s not new. It’s basically databases and middleware delivering value over a network. And to be fair, there is merit to his arguments, particularly in a world in which marketers are increasingly throwing the term “cloud” around like it’s going out of style. That said, the public’s embrace of the cloud and everything that term implies – whether you believe it is genuinely new or not – does, at least potentially, make appliances an easier sale. […]As far as I can tell, there was effectively no mention of that product. That doesn’t mean that it’s dead, but it’s omission was notable. Particularly given Ellison’s stated plans for the Sun business.[…]

[…] many of Sun’s open source engagements which don’t promise similar value – either competitive or revenue – will be scaled back or withdrawn from. Oracle, far more than Sun, is ruthlessly profit focused, and that makes certain open source commitments unlikely to be sustained.

Open Source

With Sun having invested significant resources in the past for the development of open source products and the support of open source communities, the general feels is that Oracle’s acquisition is a negative development.

Stephen O'Grady from RedMonk is not optimistic about the future of Sun technology open source communities:

There was very little discussion of open source yesterday, frankly. The word open was front and center, yes, but usually not accompanied by the word source. So from a macro perspective, I think the implications for open source communities are less than positive, given that they’re transitioning from an organization that was very pro-open source at the top to one that is, well, significantly less so. In practical terms, however, I think it’s necessary to look at the communities on a case by case basis. Java, for example, will probably be fine. Oracle’s strongly incented to keeping that community happy, for any number of reasons. MySQL, likewise, will probably be left alone for the time being. The fate of OpenSolaris, on the other hand, is less certain.

Similarly Dana Blankenhorn from ZDNet foresees a very negative impact on open source development:

Oracle now holds the copyrights on code that is basic to any open source business. Its reputation precedes it — it’s completely into profit maximization, into monopoly rents, into treating captive customers as, well, captive.[…] But this acquisition marks a turning point. It puts an end to the idea that individuals or small companies can, by organizing a community, overcome the industry’s giants. The giant just bought you out.

Also Michael Coté from RedMonk feels that Oracle will not be investing in the Open Source movement as Sun did:

Unless you have need and budget for high-performance hardware and middle-ware, it’s unclear if Oracle will care (revenue-wise) about the LAMP-stack, open source, “lesscode” crowd. That said, Ellison made a comforting comment about Java: it doesn’t have to create direct revenue, instead they’re concerned with it helping to enable revenue for the whole pie, not it’s little slice. Oracle clearly believes that its closed source offerings (Oracle DB, WebLogic, it’s apps) are “better” than open source alternatives (MySQL, Glassfish, etc.), but as long as the open source ones stay in there place, it seems they’ll be copacetic.

Sun job cuts

Regarding the major layoffs that have been going on in Sun in last years, Oracle CEO Larry Ellison has stated that in the next few months, his company plans to lay off fewer than 2,000 people, while hiring more than 2,000 people in engineering, sales and other roles. Of course He did not rule out that additional layoffs might occur later. Mr. Ellison also added that he expected Sun's chief executive, Jonathan I. Schwartz, to resign and that he hoped that Scott G. McNealy, Sun's co-founder and chairman, would stay on at Oracle, although his title and duties were not clear. Jonathan Schwartz has said in Twitter that his last blog post was “likely his last blog at Sun”.

Oracle + Sun product strategy webcasts are available online.

Those you will be visiting the Sun site, will notice that it has been changed to Oracle template and logos.

Rate this Article

Adoption
Style

BT