Q&A: Red Hat's JBoss business hits overdrive

It has been a little over two years since Red Hat acquired JBoss. Despite a relaxed public spin, rumors at the time, and for long afterwards, persisted that Red Hat didn't understand middleware, had botched the integration of the JBoss employees and culture into Red Hat, and worse.

However, in an interview Wednesday with Craig Muzilla, vice president of the Red Hat middleware business line, it became clear that JBoss--which includes all of Red Hat's middleware product line, including MetaMatrix--has finally come into its own at Red Hat. I had been hearing from different corners of Red Hat, as well as from Red Hat's competitors, that JBoss has been on a massive growth boom of late--rumors that Muzilla was happy to confirm.

JBoss is growing at twice the rate of Red Hat Enterprise Linux (RHEL). Translation? Rocket, meet bat out of hell.

The big question for Red Hat going forward is how this JBoss success will alter Red Hat's product priorities going forward. How long will Red Hat be content to be thought of as "that Linux vendor" when an ever-increasing percentage of its revenue derives from middleware?

Asay: Can you give me an update on Red Hat's middleware business?

Muzilla: We're now 26 months past the JBoss acquisition by Red Hat. Initially there were some hiccups, but we are now firing on all cylinders.

While Red Hat initially tried to use its existing RHEL sales force to sell JBoss, we learned that we needed to do some things a little differently with JBoss. So now we have dedicated sales experts for JBoss. We also went through an extensive amount of training for Red Hat's general sales team and others within Red Hat to ensure we were selling the value of JBoss.

Additionally, though Red Hat has a very strong channel business, with around 60 percent of our revenue coming through the channel, it quickly became apparent that Red Hat's existing RHEL channel wasn't right for selling JBoss. So we've beefed up our channel to work better with the type of partners that make sense to middleware, particularly system integrators.

In fact, in recognition of the need for greater understanding of and expertise in systems integration, we acquired Amentra, a leading JBoss systems integrator, earlier this year. Amentra gives us a core competency to be much more solutions driven, which has been a key to our rapidly growing JBoss business.

The result? Charlie Peters, our CFO, has disclosed that our middleware business is growing at twice the rate of our traditional RHEL operating system business.

Asay: Given the success of Red Hat in its JBoss middleware business, does this suggest Red Hat should be focusing more there, or even further up the stack?

Muzilla: Certainly people expect JBoss to be a growth engine for Red Hat. We'll continue to focus on growing this core mission of Red Hat. As [Red Hat CEO] Jim Whitehurst has stated, we have barely scratched the surface in our infrastructure business. We see a lot of promise in RHEL and JBoss.

Asay: Are there particular applications--proprietary or otherwise--that JBoss tends to get deployed alongside, similar to how Oracle was the early driver for much of Red Hat's Linux business?

Muzilla: We have a lot of commerce and transactional applications out there like Travelocity with which we work. We're strong in hospitality, financial services, government, among others, and are also seeing some interesting point of sale applications. We span horizontally across many industries.

One significant new trend, however, is how we're being considered within organizations where we've long played a relatively minor role. Five to six years ago in our Linux business, there was an inflection point when our RHEL adoption went from technical adoption of Linux to a more strategic decision to take a platform approach to RHEL. We became a platform standard within organizations.

We're at the front-end of that inflection point in our middleware business. JBoss started off as an organic, developer-driven phenomenon. We were used in departmental deployments.

Enterprises are now looking at us on a much larger, much more strategic scale. Something like 90 percent of the Fortune 2000 have JBoss somewhere in their organizations. JBoss is being used pretty much everywhere.

But now what is happening is that these organizations are looking at JBoss as a strategic platform, not simply a one-off. These conversations aren't about, "Give us support for our small project." The conversations are now, "If we're considering using JBoss as a stratetgic platform, how can Red Hat help us with this?" Twenty-four months ago we weren't having those converstions. Today, we are.

I'm sure this phenomenon isn't isolated to Red Hat. We're likely going to see other open-source software go through similar inflection points.

Asay: Fascinating. So just at the point that JBoss becomes old news in the media, it becomes hugely interesting business news to CIOs.

Muzilla: Exactly. Seven years ago you'd see CIOs awash in a number of different operating systems, but then IT organizations decided to consolidate into just a few systems, and the second alternative to whatever the dominant proprietary product tended to be open source. That strategic platform is RHEL in operating systems, and is becoming JBoss in middleware. Maybe soon it will also be an open-source CRM system, ECM system, etc.

Asay: What about the near-term product roadmap? Anything in particular that's coming down the pike soon?

Muzilla: For this I'd encourage you to read Sacha Labourey's blog on the topic. He goes into a lot of detail. A hugely significant value that we're providing in JBoss 5.0 is the ability to separate the base runtime from the middleware services from the API/programming layer. [Sacha writes:]

JBoss AS 5.0 is the first release which will give us the ability to cleanly separate those three layers. The JBoss Microcontainer abstracts us from the runtime environment and our core enterprise services have been completely componentized and aspectized so they can be fully leveraged from any higher level framework/API/language.

This new architecture means your investment in JBoss is a long term one. Our core architecture is not dependent on any fashionable spec or language du jour: personalities can be plugged in and out, a la carte, you don't have to make a bet on which API is "the" API you need, and then be locked in one of the few AS implementations that implement such API--possibly relying on weaker core middleware services.

That's the near-term vision for JBoss Application Server. Our philosophy for our next release is choice and flexibility. If you want to use Spring as a Java framework with JBoss, you will be able to do that. If you want to use our own complete stack, you can do that, too. It's up to the customer to decide, not us.

Our strategy is to focus on the application server and the application platform, but at the end of the day we need to have more components so that we have a full reference architecture. Our goal is to offer the best complete package for our customers. In some cases we may not have "The Best ESB" or some other individual component, but in terms of the overall package, we are the best.

Asay: What's your chief competitive weapon as you battle the big ecosystem vendors like Oracle that provide end-to-end software stacks?

Muzilla: Value is our primary competitive differentiator. Beyond that, we're also driving significant innovation. We are technically superior in significant ways to our proprietary competition that they simply can't match due to their closed-source code or their legacy code constraints. Is BEA rock solid? Sure. But it's lacking in other attributes--value, flexibility, innovation--and this void is driving customers to look to Red Hat.

It sounds like an exciting time to be at Red Hat. Its operating systems business continues to thrive, while its middleware business heads into overdrive. Red Hat is putting itself into a position that it could move in a number of different directions (e.g., dramatically building out its middleware business, adding applications, etc.). Success does that for a company.

Where it will end up is anyone's guess. But for now, it's great to see JBoss return to the industry as one of open source's crown jewels, rather than the wreckage of a failed acquisition.