Andrig's JBoss Blog

If You Missed JBoss World

2011-06-08T09:56:00.000-06:00

Well, if you missed JBoss World, we now have a very nice video of one of the best keynotes ever at JBoss World. It shows some of the innovation we are driving.

JBoss World 2011 Keynote

Check it out and enjoy!

Java Application Server Survey

2011-04-28T08:18:00.001-06:00

Just wanted to pass along a survey that we are doing, on which Java Application Servers you are using.

Java Application Server Survey

Its should only take a small amount of your time. It would be great to get a really big sample!

JBoss World Coming Soon!

2011-03-29T17:22:00.000-06:00

As many of you know, we have an annual conference called JBoss World. For the last couple of years this has been jointly held with Red Hat Summit. Well, this is right around the corner, and if you haven't considered attending, then you should check it out. Here is the link to the main site with all the information on sessions, speakers, etc.

JBoss World & Red Hat Summit

Each year the show has grown, and its always a good time, with a lot of things to learn about. As usual, I will be giving a performance oriented talk. In past years, I have struggled to get in all the material I want to discuss, so we have made some changes.

First, I will be co-presenting with Bill DeCoste, who is currently a solutions architect within the middle-ware business unit. Bill has been a long time JBossian and developer, and he will be discussing his experience with helping clients to prune, trim and slim down their JBoss EAP implementation, which helps things like management as well as performance. I will discuss performance tuning, as I have for the last four years now. Like always though, there are new things for us to discuss, as I have spent several months doing tests on lots of settings within the middle-ware and JVM.

The results of all of that work is not only a new presentation to give at JBoss World, but a new performance tuning guide, which will soon be a standard part of all EAP subscriptions! I'm really excited about that, and all the new things I'm going to be able to talk about.

Well, you might be thinking, especially if you have attended my performance talk in the past, well, how are we going to cover all this new material in 50 minutes?

That's the other thing that we have changed. We have a slot for an hour and a half this year, and in one of the larger rooms, so we can accommodate more people than we have in the past. It's going to be great, because we will finally have time to have a good Q&A session, which tended to happen for long after the presentation time was over in past years.

So, if your interested, and who isn't interested in getting the best performance from their applications, come to JBoss World! Here is the link to register:

See you there!

Hiring for the Performance Team

2011-03-09T07:55:00.000-07:00

I just wanted everyone to know that inside the middle-ware business unit, JBoss, we have been building a performance team. This team works upstream on projects, and on the products themselves as well, and our specific goal is to improve performance noticeably on our offerings. Besides improving performance, we have internal goals on publishing industry standard benchmarks.

So, if you know how to build Java EE applications (necessary to work on industry standard benchmarks), and you have good experience with performance tuning Java code, at all levels, then were interested in having you join. The following is a link to the job posting, where you can apply for the opening:

https://careers.redhat.com/ext/detail?redhat7016

Now, the posting says the location is London, Colorado, which is just a weird artifact of how the postings are generated, and doesn't even exist. Basically, we will hire almost anywhere in the world, so you can safely ignore the location information on the posting.

So, if your interested in a team that is doing some amazing work, apply on-line!

Java EE 6 Webinar

2010-10-08T09:01:00.000-06:00

If my last post regarding EE 6 and frameworks like Spring peaked your interest, Red Hat is having a webinar on EE 6 that you would be interested in.

Here is the landing page for our position on EE 6, and links to the webinar are on the right.

Java EE 6 for Standards Based Simplicity

For a short-cut here is a link where you can directly register for the webinar:

Java EE 6 Webinar - Java EE 6 and CDI: Moving away from clutter of Spring and other frameworks!

I think you will find this webinar informative and a must for moving to the future.

Enjoy!

Java EE 6 and Frameworks like Spring

2010-10-04T13:16:00.000-06:00

It's been a while since I posted. It's been very busy in my new role. Today, I came across this TSS thread, and thought I would share it:

Moving from Spring to Java EE 6: The Age of Frameworks is Over

Here is also a direct link to the migration article referenced in the TSS thread:

Spring to Java EE – A Migration Experience

I found the article both intriguing, and informative. The work that Red Hat has done with JSR-299, the CDI reference implementation is really becoming a water shed moment for enterprise developers. Instead of having to work around all the anti-patterns and missing pieces of Java EE, you can now count on what you need being there, and a completely extensible model for development!

There is no longer a need for external frameworks to make your life easier. That's indeed good news!

Announcing the JBoss Community Andiamo Social Group!

2010-07-01T15:38:00.000-06:00

In my relatively new role within the middleware business unit of Red Hat (JBoss of course), I am in charge of coordinating/driving our usability efforts. In that vein we have created a social group on the JBoss community site.

This new group, is a place for us to gather feedback, discuss pain points that you might have, talk about our progress and our goals.

It's just starting, although the Andiamo, or usability efforts, have been under way for awhile now. If your interested in usability of our middleware, whether from a developer or operational perspective, please join the group and contribute. Any feedback is great, positive or negative. Let's us know what you think, and where we need to improve!

The link to the community social group is as follows:

http://community.jboss.org/groups/usability

New Product Releases!

2010-03-23T10:08:00.000-06:00

As some of you may remember, back in June, 2009 we announced our "JBoss Open Choice" strategy, by which we would deliver specifically tailored products for various workloads, giving customers the maximum amount of choice for their particular needs.

Well, as of today, we have delivered on that announcement, and now have a full spectrum of products to meet your deployment requirements. So without further adieu, what does this all mean.

We now have three distinct products that you can deploy with varying levels of API support to choose from. First, we have our Enterprise Web Server (EWS) product, which is a supported Apache HTTPD and Apache Tomcat. Now, you might be thinking, doesn't Red Hat already support Apache HTTPD through Red Hat Enterprise Linux (RHEL)? The answer to that is yes, but only on RHEL. We have made Apache HTTPD available as a cross platform product now, so platforms including Windows server are supported by us. So, EWS is for those that want a very lightweight, build your own stack type of solution. Since we are talking about a very small subset of EE functionality with Tomcat. Next up, is our JBoss Enterprise Web Platform (EWP).

With EWP 5.0, announced today at EclipseCon 2010, we deliver the vision that we had for EE 6, with a web profile. It's the EE 5 technology in this case, but patterned after the web profile defined for EE 6. So, it gives developers the most often used EE 5 technologies for building web applications, but nothing more. This product will meet the needs of the majority of developers not wanting to build their own stack for web applications, without all the extra baggage for the things that you don't need. For a lot of deployments, this is the Goldilocks version, "JUST RIGHT". Of course, one size never fits all, and many people need all of the EE 5 technologies.

That brings us to the tried and true Enterprise Application Platform (EAP) 5.0, which was released last November. A complete EE 5 certified platform, that will meet the needs of any application in the enterprise.

Well, its been a great nine months since we announced the "JBoss Open Choice" initiative, and now we have delivered! If you want more information on the products, you can see our product pages on our website, using the following links:

JBoss Enterprise Application Platform

JBoss Enterprise Web Platform

JBoss Enterprise Web Server

HornetQ and SPECjms2007

2010-02-12T08:27:00.000-07:00

Today, some very interesting benchmark results have been posted. As some of you may be aware, our JBoss Messaging project, has a new community project name. It's called HornetQ. Well, some University researches have grabbed the latest community release of HornetQ, HornetQ 2.0, and ran a SPECjms2007 benchmark on that code.

Well, the results are spectacular. Today, there are only two published results for SPECjms2007, both done by the same University researchers, and both on open source projects. The first one they did was with ActiveMQ, which has a reputation for being fast. The second is now HornetQ. You can see the comparison for yourself, but we simply blow away the ActiveMQ results!

Congratulations to our HornetQ team on this awesome result!!!

You can see the results on the SPEC website here:

http://www.spec.org/jms2007/results/jms2007.html

ISV Program Hitting It's Stride

2010-02-10T13:54:00.000-07:00

One of the things that we do at Red Hat, for our supporting offerings, is work with ISV's to certify their solutions with our software. Well, in the past, for JBoss middle-ware, many ISV's would pick up the community bits and certify to those.

Now, we have an ISV program where we work with the ISV's to certify to our supported offerings. Today, Computer Associates just announced a major certification of their security products against our enterprise offerings, including Enterprise Application Platform (EAP), Enterprise SOA Platform (SOA-P), and our Enterprise Portal Platform (EPP).

This is great for customers, as they can now use Siteminder and CA SOA Security Minder with our enterprise supported platforms, with full support from CA, and JBoss, by Red Hat.

This is an awesome step forward for our mutual customers, and you can see the full press release by Computer Associates here:

CA Enables Growth of Open Source for the Enterprise

A Word About Infinispan

2009-11-30T10:11:00.000-07:00

Normally, I don't discuss the community projects in this blog, but I thought I would pass alone another blog that came out of the Devoxx conference. Infinispan is a new project in the jboss.org world, and its getting close to a community release. It really will be a game changer, in my mind, but don't take my word for it, read the following:

PearlChain Tech Blog

JBoss Middle-ware a Leader Once Again!

2009-11-09T16:58:00.000-07:00

Recently, we released our JBoss Enterprise Application Platform 5. With that release comes some very exciting technology and news.

Gartner just released their report on the EAS (Enterprise Application Servers) market, and we are once again in the leaders quadrant. Out of 28 vendors, we are only one of four to be in the leaders quadrant. The others in the leaders quandrant? Well, the usual suspects. Microsoft, Oracle and IBM. Pretty good company if you ask me. This is also the fourth consecutive year we have been in the leaders quadrant. Quite an accomplishment for us, and the only open source vendor in the leaders quadrant! If you want to see more information on Gartners report, see the following links:

Gartner's Report

Red Hat's Press Release

Red Hat's Sales Link to the Report

JBoss Enterprise Application Platform 5, is a very exciting release technology wise as well. It's our first Java EE 5 certified supported product offering. While we provided the majority of EE 5 technologies for years already, with EAP 4.3, that was not a certified EE 5 platform. This may also seem quite late to the market, but because of the capabilities we provided in EAP 4.3, our customers have already widely adopted EE 5 technologies, and now they have the complete comfort of knowing its certified too. Besides EE 5 certification, here are some of the other technology highlights:

All new Microcontainer (MC) technology, replacing the old JMX based microkernel. This new MC has been blogged about extensively already, but suffice it to say this gives us an entirely new foundation, that will allow future innovations. You can see those developing openly in projects like TorqueBox (Ruby on Rails on JBoss) and JBoss OSGi, since the MC can support OSGi directly, and consume an existing OSGi runtime as well (very flexible). This is just the beginning of a wave of innovation that will be possible with the new MC.

All new JBoss Cache integration with Hibernate, and the newest JBoss Cache, with multi-version concurrency control semantics (MVCC). Much improved speed as a second-level cache for Hibernate.

Speaking of Hibernate, we also now have Hibernate Search in the platform as a supported component. This is very exciting to me personally, as Hibernate Search solves a very common problem in applications. Being able to get fast search results from a database, where SQL queries just don't cut the mustard so-to-speak!

Speaking about fast performance, EAP 5 offers some nice improvements over EAP 4.3. In our internal testing we see much improved throughput. We also see increased scalability on clusters over 4 nodes. The clustering improvements come from improvements in JGroups and the new release of JBoss Cache that we talked about with Hibernate.

Finally, we have some exciting new capabilities in security, that Anil has blogged about fairly extensively, but I would like to highlight one thing that is near and dear to my heart. We now have a single standard mechanism for replacing all clear text passwords in configuration files. This is another big issue where compliance programs are concerned, and makes it significantly easier for customers to adopt EAP for those application workloads, and of course deepens our ability to penetrate the enterprise.

More Briefs on the Bilski Case before the US Supreme Court

2009-10-15T11:39:00.000-06:00

Groklaw has posted an article with links (Final Bilski Briefs Filed - Microsoft, Google, FFII, ABA, etc.) to some of the briefs filed in the Bilski case before the US Supreme Court. In reading through the Google brief, I was struck by this, and I quote:

"This experience only confirms the wisdom of this Court’s warning, made over 125 years ago, that indiscriminate creation of exclusive privileges tends rather to obstruct than to stimulate invention. It creates a class of speculative schemers who make it their business to watch the advancing wave of improvement, and gather its foam in the form of patented monopolies, which enable them to lay a heavy tax upon the industry of the country, without contributing anything to the real advancement of the arts. It embarrasses the honest pursuit of business with fears and apprehensions of concealed liens and unknown liabilities to lawsuits 29 and vexatious accountings for profits made in good faith.

Atlantic Works v. Brady, 107 U.S. 192, 200 (1883)."

From a case over 125 years ago, and its relevance to today is so stark!

Here's hoping the court will see the wisdom from 125 years ago, and apply to today.

More on Software Patents and the Bilski Case

2009-10-05T15:53:00.000-06:00

Today, I read the Free Software Foundation's Amicus Curiae Brief that they are submitting to the US Supreme Court in the Bilski Case.

If you have read Red Hat's brief, you might be interested in this one as well:

FSF Amicus Curiae Brief

I found it a great read, just like the Red Hat brief, and hopefully all of the brief's in support of the Bilski decision, and in support of expanding that to software will have a real impact on the Court's thinking.

Software Patents and the Bilski Case

2009-10-01T14:25:00.000-06:00

If you have been paying attention to what's happening in the courts recently, in regard to the Bilski case, then you might be interested in the following:

One Small Leap for Open Source, One Giant Leap for Mankind

The above is a blog by our CEO, Jim Whitehurst describing the steps that Red Hat is taking in this case that is going before the Supreme Court.

The following link is one from the Red Hat legal team with a little more detail:

Asking the Supreme Court to Address the Problem of Software Patents

The following link is directly to the Amicus Brief that Red Hat has submitted to the court:

Amicus Brief

I find some real hope that software patents may become a thing of the past. I sure hope so, and I'm proud to be a part of a company that is actively fighting the status quo!

JBoss World!

2009-08-20T09:46:00.003-06:00

With JBoss World just around the corner, I thought I would take some time to get the word out. JBoss World is in Chicago this year, starting September 1, and going through the 4th. It coincides with Red Hat Summit, and both events are at the same venue. Two for the price of one!

Here is the link:

JBoss World Chicago

Also, I have two sessions this year, the first being "Accelerate your JBoss", which focuses on performance tuning the JBoss Enterprise Application Platform (EAP). This year, the presentation is focused on the upcoming JBoss EAP 5, so there is a lot of things that are different. This presentation is always difficult, because the topic is so large. It's always hard to decide what to include, and what to leave out, but I think I have a good balance, and have added more stuff specific to the middle-ware, and cut out some of the stuff outside of the middleware. I also, have some better explanations on things that weren't as clear before. Overall, I think that everyone will get real benefit from this presentation.

You can see the agenda, with the abstracts here:

Agenda (abstracts)

I will also be taking part in a BOF for application platforms, which should be great. It's earlier than the last JBoss World as well, so it should be a lot of fun.

If your coming, I look forward to meeting any and all of you that I can. I will be working in the booth from time to time as well.

If your not currently planning on coming, please consider it. I know travel budgets may be tight at companies right now, but this will be well worth the money.

Hope to see you there!

JBoss EAP 4.3 is Common Criteria Certified

2009-07-07T07:36:00.003-06:00

It's been a while since I blogged about anything here, and I'm tardy on this one too. JBoss has been gaining momentum in the government space, but everyone that has any experience with working with our federal government knows, you can only go so far unless you go through the common criteria certification process, and get certified.

Well, I'm proud to say that JBoss EAP 4.3 is common critieria certified! This was a huge step forward for us, and a special thanks goes out to Anil Saldhana for leading this effort. Here is Anil's blog post on the subject as well:

JBoss EAP 4.3 is Common Criteria Certified

So, if your in the federal government space, and interested in our technologies, you now have a new choice of common criteria certified software for your use!

Yet More Reasons to Subscribe to our Enterprise Platforms

2009-02-12T08:52:00.003-07:00

I wanted to return to this topic again, because it is a continual discussion that we have with potential customers, and I realized that there is another large reason for subscribing that I haven't talked about in the past.

In the last several months, security has come up many times internally and externally. With our JBoss.org software, we are really targeting making it very easy for developers to get up to speed quickly, and having everything locked down flies in the face of that goal.

With that in mind, what do we do in our productization process for security?

First, we have dedicated resources internal to Red Hat that handle security of the JBoss Enterprise Platforms. These resources do a review of our platforms from a security perspective, and we address any issues they find through configuration and bug fixes. This is a pretty standard process for most vendors, and we are no different.

Second, we lock down all the management and exposed interfaces. So, for example, the JMX console has no authentication requirements within the JBoss.org Application Server release. Within the EAP, the JMX console is secured with a username and password. While this is a simple example, it is very, very important. We find people with production web sites running on JBoss.org AS all the time with a completely unsecured JMX console. Now what does this mean?

Well, it means I can change the configuration of the running AS, undeploy the applications, and even shut it down completely, from anywhere in the world, without anyone knowing what happened! Ouch!

So, between the bug fixes that are addressed for security, and the locked down configuration, out-of-the-box, we add significant value to the subscription, and is yet another reason why you should subscribe.

JBoss Division of Red Hat Gives Birth!

2009-02-11T13:43:00.004-07:00

Well, you might be asking yourself, what is this title all about? JBoss has given birth? What's that all about?! Well, it's about our acquisition of Metamatrix.

As some of you may know, we acquired a company by the name of Metamatrix, and Metamatrix is all about data federation. Metamatrix's technology was not open source at the time of the acquisition, and while we have been selling subscriptions to it, just like our subscriptions on our open source technology, it wasn't open source. Well, I'm proud to announce the birth of a new open source project into the JBoss family of middle-ware, and its called Teiid!

Yes, that's right, the new project is the culmination of a tremendous amount of work, by a great team of developers to open source the Metamatrix technology. We always said that we would open source the technology, and maybe some folks were beginning to doubt us, but we always live up to what we say we are going to do in the open source community, and its now officially here. When you have as large a code base as Metamatrix has, and you have some encumbered code that has to be replaced, it is a significant effort. All this had to be done, while simultaneously deliverying new functionality and patches to existing and new subscription customers. It's been a long road for the team, but its delivered! This will become the basis for all future products, which will be our Enterprise Data Services Platform.

I want to congratulate the entire team on mission accomplished!

You can find the new project, and the Eclipse based tools for it at the following links:

Teiid
Teiid Designer

Go download it, test it, look at the code, and get involved, even if its just feedback in the forums and mailing lists.

Enjoy!

JBoss Virtual Experience 2009!

2009-01-21T08:10:00.003-07:00

I wanted to take a moment to let everyone know that we at JBoss are doing a virtual trade show, called JBoss Virtual Experience 2009!

I would encourage everyone to register and attend this on-line event. It's coming up on February 11th, and since its an on-line event, there's no travel required, just your time. And as we all know, in these trying economic times, travel is being cut back at probably all companies these days.

What I'm most excited about is that there will be a live chat. I will personally be involved in that, and some of our open source project leaders will be involved as well, so this is your chance to ask questions live of some open source community leaders and JBoss division leaders. Please join us, I think you will enjoy the event, and get a lot of good information from it. I have included a link to the web site above, so go there, register, block your day off, and join us!

It's Official

2008-12-11T12:31:00.003-07:00

Not to long ago, I blogged about passing the Java EE 5 TCK. Well, due to some issues with interpretation of the TCK license agreement, we didn't show up on the Java EE compatible page from Sun. Well, we rectified the issue to Sun's satisfaction, and with our GA release of AS 5, its now official.

We are certified Java EE 5 compliant!

You can see the proof here:

Java EE 5 Compatible Implementations

and you can go get JBoss AS 5 from here:

JBoss.org AS Downloads

Try it out. I think you will like it!

JBoss Innovation Engine Heats Up!

2008-11-19T17:53:00.002-07:00

It's not often I talk about the individual projects at jboss.org, but in this case, I think the latest release of JBoss Cache deserves special mention. When you consider that our currently shipping EAP 4.3 release includes JBoss Cache 1.4, and the JBoss Cache developers just went GA in the community with their 3.0 release, the innovation engine is really heating up.

We released EAP 4.3 on January 11, 2008 (if memory serves me right), and in the intervening 10 months the JBoss Cache developers have release JBoss Cache 2.0, 2.2, and now 3.0. That's right, two major releases and a minor in ten months. This doesn't include any bug fix releases, and support of the currently shipping version! An awesome job, and its just another example of the strength of the open source development model. One of the other things that deserves special mention, is that JBoss Cache implements a new concurrency model called MVCC or multi-version concurrency control. This model provides for a transactional cache that can scale very well. The JBoss Cache team has done some extensive benchmarking, and the results are spectacular! Finally, because they have moved so fast, we have been able to include JBoss Cache 3.0 in the upcoming community GA release of AS 5! This in turn will become the release in our upcoming EAP 5!

Finally, for all the gory details, I'll point you to the JBoss Cache blog post announcing the release, and it has links to everything you need.

http://jbosscache.blogspot.com/2008/11/jboss-cache-300-goes-ga.html

Enjoy!

My Dream for OpenJDK

2008-11-17T16:13:00.002-07:00

When Sun open sourced the JDK (finally a true reality), I was very pleased indeed. It was something the open source community wanted, and Java needed to continue to be relevant in the future. Having said that, I have seen some very positive things happening, much of which has been driven by the Red Hat OpenJDK team.

That team has created a complete open source tool chain to build and package OpenJDK for Linux. We are seeing all the Linux distributions rally around this, and it is very encouraging indeed. Having an open source JDK that is built completely with open source tools, and packaged for all the major Linux distributions is a great thing. It gives developers a platform that they can count on. While this is a great step forward, it isn't the total picture that I envisioned for OpenJDK.

I would really love to see other operating system vendors embrace OpenJDK for their respective platforms as well. Seeing the same base used across all operating system platforms will give developers what Java promised all along. Write once, run anywhere could truly be a reality. Having all operating system platform vendors embrace OpenJDK will yield benefits to developers in other ways besides consistency.

If everyone starts contributing to the same code base, then the network affect of innovation can take off in the OpenJDK community, and we can see real improvements across the board. Whether that be in innovations around garbage collection, innovations around multi-core support, you name it. All the best minds working together, versus working separately, will benefit all the vendors customers, and shouldn't hurt any one vendor either. A rising tide lifts all boats.

So, my dream is that all the operating system platform vendors would bury the hatchet on any past issues they may have had with Sun, and join the OpenJDK community. I would even like to see the open source community working on other JDK implementations start working in the OpenJDK community. We will all benefit!

Our Continued Commitment to What Our Customers Want

2008-11-17T15:32:00.002-07:00

It's been awhile since I blogged. Things have been pretty busy around here. My first couple of blog posts were centered around how we create our enterprise platforms, and one of the things I covered was the combinations of operating system and JVMs, along with databases that we test with. At the start, it was already a pretty good list. Well, in the intervening months since then, we have added considerably more operating system/JVM combinations, and new databases, and I wanted to highlight that here.

With our latest cumulative patch release we have now certified not only the Sun 1.6 JVM, but now also the Oracle (formerly BEA) JRockit 1.6 JVM, HP's HP-UX 1.6 JVM and IBM's 1.6 JVM. The specific versions are as follows:

Oracle JRockit 1.6.0_05
HP-UX JDK 1.6.0_01
IBM JDK 1.6.0 SR1

We have also tested with two versions of IBM's DB2 database. Some of our Big Blue customers really needed this, and we have delivered. The specific database versions and JDBC drivers are as follows:

IBM DB2 8.2.7 - Universal JDBC Driver v2.10.52
IBM DB2 9.1 Fixpack 3 - Universal JDBC Driver v.3.1.57

We now have some 54 (44 OS/JVM combinations and 9 database/JDBC Driver combinations) tested combinations in our matrix of certified operating systems, JVMs and databases. While I haven't checked this against our competition, I believe we may have one of the best platform coverages of any middle-ware vendor today. Here is the full list of certified combinations:

Red Hat Enterprise Linux v5 (latest update) - x86 - Sun JDK 1.6.0_07
Red Hat Enterprise Linux v5 (latest update) - x86 - Sun JDK 1.5.0_15
Red Hat Enterprise Linux v5 (latest update) - x86 - Oracle JRockit 1.6.0_05
Red Hat Enterprise Linux v5 (latest update) - x86 - Oracle JRockit 1.5.0_08
Red Hat Enterprise Linux v5 (latest update) - x86 - IBM JDK 1.6.0 SR1
Red Hat Enterprise Linux v5 (latest update) - x86 - IBM JDK 1.5.0 SR6
Red Hat Enterprise Linux v5 (latest update) - x86_64 - Sun JDK 1.6.0_07
Red Hat Enterprise Linux v5 (latest update) - x86_64 - Sun JDK 1.5.0_15
Red Hat Enterprise Linux v5 (latest update) - x86_64 - Oracle JRockit 1.6.0_05
Red Hat Enterprise Linux v5 (latest update) - x86_64 - Oracle JRockit 1.5.0_08
Red Hat Enterprise Linux v5 (latest update) - x86_64 - IBM JDK 1.6.0 SR1
Red Hat Enterprise Linux v5 (latest update) - x86_64 - IBM JDK 1.5.0 SR6
Red Hat Enterprise Linux v5 (latest update) - System Z s390x (64-bit) - IBM JDK 1.5.0 SR5
Red Hat Enterprise Linux v4 (latest update) - x86 - Sun JDK 1.6.0_07
Red Hat Enterprise Linux v4 (latest update) - x86 - Sun JDK 1.5.0_15
Red Hat Enterprise Linux v4 (latest update) - x86 - Oracle JRockit 1.6.0_05
Red Hat Enterprise Linux v4 (latest update) - x86 - Oracle JRockit 1.5.0_08
Red Hat Enterprise Linux v4 (latest update) - x86 - IBM JDK 1.6.0 SR1
Red Hat Enterprise Linux v4 (latest update) - x86 - IBM JDK 1.5.0 SR6
Red Hat Enterprise Linux v4 (latest update) - x86_64 - Sun JDK 1.6.0_07
Red Hat Enterprise Linux v4 (latest update) - x86_64 - Sun JDK 1.5.0_15
Red Hat Enterprise Linux v4 (latest update) - x86_64 - Oracle JRockit 1.6.0_05
Red Hat Enterprise Linux v4 (latest update) - x86_64 - Oracle JRockit 1.5.0_08
Red Hat Enterprise Linux v4 (latest update) - x86_64 - IBM JDK 1.6.0 SR1
Red Hat Enterprise Linux v4 (latest update) - x86_64 - IBM JDK 1.5.0 SR6
Red Hat Enterprise Linux v4 (latest update) - Azul - Azul JDK 1.5.0_11
Red Hat Enterprise Linux v4 (latest update) - System Z s390x (64-bit) - IBM JDK 1.5.0 SR5
Red Hat Enterprise Linux v4 (latest update) - System Z s390x (31-bit) - IBM JDK 1.5.0 SR5
Microsoft Windows 2003 - x86 - Sun JDK 1.6.0_07
Microsoft Windows 2003 - x86 - Sun JDK 1.5.0_15
Microsoft Windows 2003 - x86 - Oracle JRockit 1.5.0_10
Microsoft Windows 2003 - x86_64 - Sun JDK 1.6.0_07
Microsoft Windows 2003 - x86_64 - Sun JDK 1.5.0_10
Microsoft Windows 2003 - X86_64 - Oracle JRockit 1.5.0_08
Sun Solaris 10 - Sparc - Sun JDK 1.6.0_07
Sun Solaris 10 - Sparc - Sun JDK 1.5.0_15
Sun Solaris 10 - x86 - Sun JDK 1.6.0_07
Sun Solaris 10 - x86 - Sun JDK 1.5.0_15
Sun Solaris 9 - Sparc - Sun JDK 1.6.0_07
Sun Solaris 9 - Sparc - Sun JDK 1.5.0_15
Hewlett-Packard HP-UX i2 - PA-RISC - HP JDK 1.6.0_01
Hewlett-Packard HP-UX i2 - PA-RISC - HP JDK 1.5.0_06
Hewlett-Packard HP-UX i2 - ia64 - HP JDK 1.6.0_01
Hewlett-Packard HP-UX i2 - ia64 - HP JDK 1.5.0_06

For databases:

IBM DB2 9.1 Fixpack 3 - IBM DB2 Universal JDBC Driver v3.1.57
IBM DB2 8.2.7 - IBM DB2 Universal JDBC Driver v2.10.52
Oracle 10g R2 - Oracle JDBC driver v10.2.0.1
Oracle 10g R2 - Oracle JDBC driver v10.2.0.2
Oracle 9i - Oracle JDBC driver v10.2.0.1
Oracle 9i - Oracle JDBC driver v10.2.0.2
Microsoft SQL Server 2005 - Microsoft SQL Server 2005 driver v1.1.1501.101
MySQL 5 - mysql-connector-java v5.0.7
PostgreSQL 8.1 - PostgreSQL v8.2 JDBC3 with SSL (build 504)

In the future look, for even more news on the database front. This picture will continue to get better and better over time, and we even see increasing interest from the vendors to have us do the certification testing, along with the interest from customers.

Happy computing!

Mobicents is SIP Servlet 1.1 Compliant!

2008-10-01T09:22:00.002-06:00

Just recently our AS 5 CR 2 release went out, and it passes the Java EE 5 TCK, and now the hits just keep on coming.

Our Mobicents project is now the first open source project that is JSR-289 compliant, which is the SIP (Session Initiation Protocol) Servlet 1.1 specification.

Congratulations goes out to the entire Mobicents team! Great job. Besides being the first open source JSR-298 compliant container, Mobicents is also JSLEE 1.0 compliant as well, bringing the entire range of telecommunications specifications to bear.

If you are interested in more information about Mobicents, and our JBoss Communications Platform based on Mobicents, then see Jean Deruelle's blog post on the new SIP Servlets container, the Mobicents project page on JBoss.org, and finally the JBoss Communications Platform page on the Red Hat website.

AS 5 CR2 is EE 5 Certified!

2008-09-18T12:39:00.002-06:00

Today we released our AS 5 CR2 release, and we have now passed the Java EE 5 TCK! This is a huge step forward for us, and I would love to see everyone download and really hit this release hard. It's our last CR prior to going GA, and we would like to encourage everyone to hammer this release, so we can address any final issues that we may not be aware of.

You can find the release on jboss.org here.

Here are the release notes.

Hammer it, and give us your feedback!

Enjoy!

Seam 2.0.x Support in our Enterprise Application Platform!

2008-09-09T15:17:00.003-06:00

Customers asked for it, and we listened. As of today, we have released what we are calling a feature pack on top of EAP 4.3 CP02 (Cumulative Patch Release 2). Sometimes the community gets way ahead of our product, which is the way its supposed to be, but in the case of Seam, customers adoption rates has been very high. With that, we probably had fifty percent of all customer support issues related to Seam based on the Seam 2.0.x code base, instead of the 1.2.1 code base that is actually in the shipping product!

So, a lot of customer adoption and demand, and some very specific cases with customers asking us to fully support Seam 2.0 within the product. With that, we had to make a decision. Do we create an entire new release of EAP, such as an EAP 4.4, or do we come up with a way to get customers Seam 2.0 support on the current release? In looking at those options, it made the most sense for us as a business, and for customers, that we support Seam 2.0 on the currently shipping EAP 4.3 release. Thus the "Feature Pack" was born!

The concept of the "Feature Pack" is simple. We would provide additional features on top of an existing release, maintaining that releases backward compatibility and maintaining the support life-cycle, but provide what customers would normally have to wait for in a minor or major release such as our planned EAP 5 release! In many cases, this just is not possible for some customers, and with Seam 2.0 there are so many compelling features over Seam 1.2.1, that customers really needed to move to that. The catch, of course, is that it was only a community release, and not shipped with our product.

Well, we have taken care of the issues, and customers can now download and apply the feature pack to their existing installations. The feature pack installs Seam 2.0.x in a new directory structure, side-by-side with the older Seam releases, so if you have any Seam applications built with Seam 1.2.1, they will continue to function as they did before, but you can now migrate to Seam 2.0, and get all the new capabilities that come with Seam 2.0! This is exciting news, and we also already have tooling support built into JBDS (JBoss Developer Studio) 1.1!

So, go and download the feature pack, and have fun with all the new capabilities of Seam 2!

Red Hat IT Switches to JBoss EAP 4.3!

2008-08-27T14:32:00.004-06:00

I must first say congratulations to the Red Hat IT department! They did a wonderful job with the conversion to JBoss Enterprise Application Platform 4.3! Next, a little history for the uninitiated.

Red Hat's IT department has built many applications that are used in the day-to-day operations of Red Hat. Prior to the JBoss acquistion, they used a lot of standalone Tomcat, with some in-house customizations I am told. Of course, like a lot of IT shops, these applications grew organically based on need, and there really wasn't an organization dedicated to creating a standard infrastructure and architecture for all applications. Recently, a new architecture organization was born, and led by Matt Hicks. This organization is the one responsible for pulling together a standard infrastructure and architecture for existing as well as new application efforts within Red Hat. This is a wonderful development from the middle-ware divisions perspective, as it gives us a partner to work with internally. It has also spawned a completely new open source project around virtualization and provisioning that is quite impressive! You can see the resulting Genome project at:

Genome

With that, Red Hat IT has moved all the customer facing web applications from the old Tomcat infrastructure to the new JBoss EAP 4.3 infrastructure. With this move, they now have a completely standard infrastructure and architecture that can be replicated through the Genome work in a moments notice. So, what were some of the benefits?

Well, first, they are leveraging the clustering capabilities within JBoss EAP, and have a highly-available deployment, that also scales very well across the cluster. Second, they were able to achieve significant performance improvements. They have been able to reduce the hardware that these applications were running on by 50%, and still have more head room for growth than under the old Tomcat infrastructure. This makes for an incredible savings in hardware, power and cooling costs!

I would also like to give a special thanks to the project lead, Chris Duryee. Having worked in IT departments over the course of the first 21 years of my career, and I can really appreciate all the hard work, and diligence these types of conversions take!

Of course, this is not the end of the story for Red Hat's IT, but just the beginning. In the future, they will also be leveraging our Enterprise SOA platform, and we look forward to the amazing results of that effort!

Congratulations to the entire team!

JBoss ESB Wins!

2008-08-05T15:29:00.003-06:00

When you consider that our JBoss ESB project is not very old, and we introduced the first product, JBoss Enterprise SOA Platform, based on it only in February of this year, it is a real honor to win InfoWorld's Bossie Award in the open source platforms and middleware category.

We may have been a little late to market, but the solution is strong.

See the following link to the InfoWorld website:

Best of open source platforms and middleware - Enterprise Service Bus

Congratulations to the team here at JBoss working on JBoss ESB!

What's Wrong with the SPECjAppServer2004 Benchmark?

2008-07-23T16:01:00.010-06:00

Recently, I have noticed that companies are asking about SPECjAppServer2004 results for our Enterprise Application Platform. While this isn't entirely new, it seems like it is happening more. In most cases prospective customers don't actually understand that what they are asking for isn't necessarily relevant to their environment. So, exactly what's wrong with this benchmark?

Well, before answering that question directly, let's first talk about benchmarks in general. Industry standard benchmarks are created through a consensus process within the respective organization. They involve the vendors that all have a stake in the outcome, so they all bring their agenda that includes their products strengths (they also try to avoid their products weaknesses). Of course, no one vendor gets everything they want, but they are not independently created without regard for any individual vendors products. The vendors have influence, and that skews the benchmark from the beginning. Besides the vendor influences in the creation of benchmarks, the actual implementation of the benchmarks is not very realistic to a business application. At least not any business application that I have ever written or seen.

I was in IT, developing custom business applications for over 21 years, before joining JBoss (now the middle-ware business unit of Red Hat), and I can tell you from experience, that benchmarks do not reflect real world business applications. In fact, they rarely have anything but trivial business logic in them. They also don't reflect the technology that gets used by customers. They reflect the technology that is either in a specification, or the technology that particular vendors would like to push. This is especially true, when the vendors don't have other alternatives of their own.

The other problem with benchmarks, in general, is that the numbers that any vendor creates with them will not translate directly to anything meaningful in your business. Is something that holds no direct meaning to your business a criteria that you should use in making a decision?

After contemplating my last, albeit rhetorical question, let's get back to the initial question. So, what is wrong with the SPECjAppServer2004 benchmark?

The SPECjAppServer2004 benchmark, suffers from all the ills that almost all benchmarks suffer from. First, it looks a lot like TPC-C, as it appears to be modeled after it, but written in Java, using J2EE 1.4 technologies. So, its not a realistic representation of a business application. Second, its business logic is trivial, and in no way compares to the typical complexities of business logic in real world applications.

All the real-world applications that I have worked on had millions of lines of business logic, along with millions of lines of code that were more technical in nature, for interfaces to other systems, persistence, transaction processing, etc. Third, it is clear that the author, or authors, of this benchmark have never written a business application in their lives. Or if they had, it would have been a very poor one indeed. The code uses floating point numbers to represent dollars and cents! Ouch! Real world business applications written in Java would use the BigDecimal class to represent dollars and cents. Using native types certainly will make the benchmark faster, than using BigDecimal, which uses arbitrary precision arithmetic, but it will create results that aren't correct. The more complex the real world calculations are in your business application, the larger the calculations errors will be. For example, I developed an application where there was complex discounting schemes with discount percentages carried out to four decimal places. The calculation had to be applied back to the original charges (not just a total), and in doing that you had to go back over every detail charge (could be millions and millions per account). In order to do that you had to truncate values, and roll the remainder down, and only round at the end, so that it came out correctly, and no matter what kind of slicing and dicing you did, reporting wise, the totals would always foot. You simply cannot do those kind of calculations with floating point numbers with any accuracy at all! Finally, the benchmark is heavily dependent on EJB 2.x Container Managed Persistence (CMP).

This is the perfect example of a benchmark utilizing technology that is not relevant to customers. With the extreme limitations of EJB QL for EJB 2.x CMP, there aren't many real world applications that can use CMP. In our own customer based the vast majority have turned to alternative ORM technology like Hibernate. We also know that this is true in WebSphere and Weblogic shops as well. I think its also illustrated by the fact that BEA (years before the acquisition by Oracle), announced support of Hibernate with Weblogic! Do you think they would do that, if their customers were using CMP? Just like us, their customer base turned their back on EJB 2.x entity beans with CMP a long time ago.

When you wrap all these things up, what value does a SPECjAppServer2004 result actually provide? Well, I think we can safely say that it doesn't provide any value. In my experience, the best thing that any customer can do is run their own application against the various middle-ware platforms, and compare those results. It is what I always did, when I was in IT. Industry standard benchmarks might be a tempting short-cut, but in this case, it really isn't going to tell you anything meaningful.

There simply is no substitute from seeing your own workload running on the potential solutions!

Thinking about Value with an Enterprise Platform Subscription

2008-07-11T09:46:00.003-06:00

Recently, I have been contemplating the way people view our subscriptions for the enterprise platforms, and what they think the value proposition is. In fact, there was a recent comment, that really made me think. That comment was about cost per incident. In thinking this way, the customer was wondering what value they were getting from the subscription when their cost per incident was so high.

In dealing with that comment directly, I think this is a totally flawed way of thinking about the value of a subscription. The subscription costs are already low, when you compare it to our competitors license costs plus maintenance costs, or even their maintenance costs alone. Having said that, if you have paid for a large deployment, and you don't have very many incidents, and in this context, we are talking about the number of times that the customer has to engage our support. Then the cost per incident could be quite high. Now, why is this way of looking at value flawed?

Well, you are basically arguing for deploying low quality software in your production environment! Is that really what customers want? If they were asked this question directly, they would say no. So, cost per incident is not a good way to look at whether you are paying too much for a subscription, or whether you should subscribe at all, versus using the jboss.org bits for free. So, this leads us what is the value of a subscription for the enterprise platforms.

For me personally, when I deployed infrastructure software into the enterprise (I was in IT for 21 years prior to joining JBoss), my first concern was stability of that infrastructure. I didn't want to constantly be dealing with failures in my operating system, middle-ware, databases, etc. If you will, I wanted my incident rate to be very low. When I did have a problem, I wanted to be able to solve that problem in the shortest amount of time possible. So, cost was not my first concern, quality was.

With our enterprise platforms, quality is second to none. When you consider the network affect of the open source development model, plus the hardening process that we go through with the enterprise platforms, the result is software with unparallel quality. This results in production deployments with low incident rates. I like to think about this cost per incident, not in the sense of how much is being paid for the subscription divided by the number of incidents, but by the lost opportunity cost of having my internal staff working on middle-ware issues, versus working on something that might actually deliver value from a business perspective. This is what Geoffrey A. Moore, author of "Living on the Fault Line, Managing for Shareholder Value in ANY Economy", would call core versus context.

If you ask yourself a simple question about what is core to your business versus what is context to your business, I would expect that virtually everybody would say that middle-ware support is not core to their business, its context. It might be mission critical context, don't get me wrong, but it is still context nontheless. So, should you really be spending the brain power of the organization on figuring out problems with middle-ware, or building that next great feature or application that generates more revenue?

Therefore, the value proposition is both in the quality of the support given, and the quality of the software delivered. When I deployed JBoss in mission critical environments, I also experienced superior support from the JBoss support organization. We never had an issue that went unresolved in five years of production usage (prior to me leaving), I couldn't say that about any other vendor we worked with. While this was my own personal experience, and is certainly a microcosm of all customer experiences, it was one of the reasons I felt very comfortable joining an organization in making such a big career change.

Remember the value of the subscription is not just in the support but in the quality of the software as well. You don't have to pay a big upfront license cost to get value, and in fact I would argue why should the customer absorb all that risk. The vendor that produces and supports that software should be taking that risk, not you the customer!

More Reasons to Adopt JBoss Enterprise Platforms

2008-06-05T12:12:00.003-06:00

In my last post, I talked about some reasons to adopt JBoss enterprise platforms, versus our jboss.org software. My main focus was on the additional testing that we do for the software, and I mentioned the stability and support as an aside. In this post, I would like to address the issue of stability and support.

So, what do you get as a subscriber in terms of stability and support? Let's start by addressing stability. For a lot of customers, stability is very important. They make a large investment in a custom developed application, and it is expected to be used in production for many years. Some companies capitalize this work, from an accounting perspective, and then amortize it over three, five or even seven years! Over that time period, even if they are doing regular upgrade releases to the software, they need the middle-ware to be as stable as possible. API and even configuration incompatibilities, can cause major problems. Typical cycles for upgrading middle-ware in an enterprise setting can also take six to nine months, or longer! With these types of cycle times in mind, our products have to be stable.

So, just how do we keep the products stable? We have instituted a quarterly cumulative patch process within the enterprise platforms. On a quarterly basis, we take defects and fix them and roll them together in a release (no new features). Some of these defects might have been delivered to customers prior to the quarterly cycle as hot fixes, depending on severity. We leverage the test suites that I detailed in my earlier post, and make sure there are no regressions. Also, the defects come from our customers, as a priority, but we may also fix issues that we know affect the enterprise platform that may have been caught in our upstream jboss.org releases. This gives us a measure of additional stability, that even though a customer has not seen this issue, we have proactively fixed it knowing that it will probably crop up down the road. Of course, this is the real power of an open source model. You have the widest and most varied deployments to flush out issues, and address them proactively. This cumulative patch process goes on for three full years, where customers can be assured that if they upgrade to any of these releases their applications will not break. For another two years, we will still produce these cumulative patch releases, but they will contain only security errata. This is vastly different than the jboss.org releases that we used to support in the old model, where an upgrade from one release to another might have feature enhancements, configuration changes, and occasionally API changes between dot dot dot releases, making it very difficult for customers. So, what does this mean for support?

Without going into details of exactly how we structure our support organization (there are better people than me to explain that), the model of supplying cumulative patch releases, and not changing API's and configuration, makes the support engineers jobs a lot easier. This, in turn, leads to much higher quality support. With this model, the support engineers don't have to wonder what possible combination of components have been thrown together. They don't have to worry about configuration or API changes that may be different from customer to customer. This all makes reproducing and fixing issues much simpler. In fact, its a win-win situation for us as a company and for the customer. It allows us to scale our business much more affectively, and at the same time delivers a much better customer experience. Customers don't have to worry about support quality suffering as we grow (and we are growing rapidly).

In summary, stability and quality support is important to our customers, and we deliver that better than any other vendor! I know this from personal experience as well, as I was a JBoss customer prior to joining the company.

Why Adopt JBoss Enterprise Platforms?

2008-04-18T12:40:00.006-06:00

Since we started releasing our Enterprise Platform offerings, there has been some confusion as to why would someone become a subscriber and adopt the enterprise platforms, such as the Enterprise Application Platform or the Enterprise SOA Platform, versus just downloading the jboss.org bits and using them for free?

This is certainly a legitimate question, and one that I get a lot. It also has some history behind it, since our original product model was every release on jboss.org was a supported product that you could purchase a support subscription for. With this split between having jboss.org releases be just for community consumption, and not for paying customers, confusion has reigned supreme. With that, let's talk about the reason we even went down this path.

One of the biggest issues our customers faced was stability. Under the old product model, there were fairly major changes and/or feature enhancements even in dot dot dot releases, which had the potential for breaking customers applications. That was certainly the number one problem our customers were facing, and we had to address that. While there are other issues that drove us to the new product model, stability was the number one reason. Many of our customers want a product that they can deploy, and that they know will guarantee compatibility over many years. We just didn't deliver that under the old model.

With that in mind, what are the differences between the enterprise platform releases and the jboss.org releases?

I will use our JBoss Application server, and Enterprise Application Platform as an example. With the JBoss AS project, we release based on passing two test suites (unit tests and compatibility matrix tests) which are a part of the open source project. When we have the features and fixes targeted for that release, and the test suites are 100% passing, we release the code to the community. Now, the test suites are substantial, with getting close to 4,000 tests in them. Having said that, it is run with one JVM, the Sun 1.5 JVM, on one OS platform, which happens to be Red Hat Enterprise Linux (this has actually been the case for many years, even before the Red Hat acquisition of JBoss). Now, let's contrast that to what we do for the Enterprise Application Platform.

We still do use the aforementioned test suites, but we run them on a combination of JVM's and OS's. For example, with EAP 4.3, we have run the test suites against the following JVM and OS combinations:

BEA JRockit 1.5 JVM on RHEL 4 x86
BEA JRockit 1.5 JVM on RHEL 4 x86_64
BEA JRockit 1.5 JVM on RHEL 5 x86
BEA JRockit 1.5 JVM on RHEL 5 x86_64
BEA JRockit 1.5 JVM on Windows 2003 x86
BEA JRockit 1.5 JVM on Windows 2003 x86_64
HP 1.5 JVM on HP-UX 11i IA64
HP 1.5 JVM on HP-UX 11i PA-RISC
Sun 1.5 JVM on RHEL 5 x86
Sun 1.5 JVM on RHEL 4 x86_64
Sun 1.5 JVM on Solaris 9
Sun 1.5 JVM on Solaris 10
Sun 1.5 JVM on Windows 2003 x86
Sun 1.5 JVM on Windows 2003 x86_64
Sun 1.5 JVM on RHEL 4 x86
Sun 1.5 JVM on RHEL 4 x86_64

This is our tier 1 set of platforms, and post the release we have continued to test other combinations, such as the IBM 1.5 JVM (EAP 4.3 was released in January, 2008). Through this testing we catch issues both with the test suite, and with our code that never makes it to our customers. Besides this testing, we also incorporate some of the individual component test suites, such as:

EJB 3
JBoss Web Services
JBoss Remoting
Hibernate

This in and of itself is a big step forward in our quality, but that isn't were we stop.

We have created another set of tests for performance and scalability. We do performance and scalability testing for:

EJB 3
JBoss Seam

These are our two most important parts of the Enterprise Application Platform, from a performance perspective, for our customers. Therefore, we want to make sure that there aren't any inherent performance and scalability problems with these features of the platform. We also use these tests for reliability testing, in the form of a soak test.

In the soak testing, we run at high-load, for an extended period of time (24 to 36 hours). This allows us to make sure there aren't issues that our customers will face that affect uptime of their deployments. Once again, we don't stop there.

With Hibernate we test against the following set of databases:

Microsoft SQL Server 2005
MySQL 5.0
PostgreSQL 8.2
Oracle 9i
Oracle 10g

Against the above databases, we also use the database portion of the J2EE 1.4 TCK to make sure our EJB 2.x implementation works appropriately across these databases as well. This is also our tier 1 list of databases, just like the JVM and OS combinations. Since release, we are also doing testing with DB2 and Sybase. Again, we aren't done yet.

We also do testing around our clustering technologies, such as:

HTTP Session Replication
JMS Clustering
JMS Fail-over

With the introduction of JBoss Messaging in the EAP 4.3 release, we now have clustering and fail-over capabilities that were not in the old JBoss MQ that it replaces, and hence we have added testing specifically for these new capabilities. Finally, there are a set of tests that cover installation and management, which include:

GUI Installer
JBoss Operations Network (JON)
RPM testing

Whew! As you can see this is a lot of testing that is new with the Enterprise Application Platform, that never existing with our old product model, and is only applied to the platform products. Now what do we do with all the fixes that result from all this new testing?

Fixes go directly onto the appropriate branch for the enterprise platform, and they also go upstream into the community. Of course, the fixes that go upstream don't have any SLA associated with them and the code branches are substantially different from each other. Also, the upstream jboss.org releases may also have new features in them that aren't in the enterprise platforms. Post release of the enterprise platforms, we also do quarterly cumulative patch releases, and over time, because we don't add new features, the code continues to diverge from the upstream jboss.org releases. This goes right to the heart of the stability our customers want, and we deliver these platforms with a quality level that is unmatched in our old model.

If you want a stable platform to develop your applications on, that won't break your applications over time (3 year full support, with 2 year of security errata support), then the JBoss Enterprise Platforms are for you!