InformationWeek's Cloud Computing Destination

VMware's Cunning Acquisition: SpringSource

Cloud computing and virtualization function hand in glove. We knew that. What we didn't know was that there are likely to be efficiencies if the application is built from the ground up for the cloud. The Spring Framework is one of those new development platforms that make it easier to develop Java applications--for the cloud.

In buying SpringSource, VMware may be correct in assuming that, in the long run, frameworks will serve as one of the springboards to cloud computing. And that means VMware has stopped thinking as a virtualization vendor and started thinking as a cloud supplier.

I suspect Rod Johnson, the founder of the Spring Framework project and CEO of SpringSource, has made a similar transformation.

He is a classic open source leader, open to new contributors and input but centered on a core set of ideas that let other developers crowd around and work alongside him, with minimal friction and without a loss of focus. During the VMware investor relations call Monday, it seemed at moments as if VMware's main goal in buying SpringSource was to get an innovative process associated with Rod and the developers working with him working inside its doors.

"We are doing this for the potential for innovation here," said Maritz during the call, without specifying exactly what innovation was expected. VMware quietly invested $2 million in SpringSource nine months ago to get a board seat and more closely observe the company. Apparently, it liked what it saw. Yesterday it said it had decided to invest $362 million more.

With the acquisition, "we can do a remarkable set of things for applications," details still to come, said Maritz.

Doesn't that draw VMware into direct competition with partners who currently think they are providing application management systems?

"We're talking about innovation, about doing things that no one else is doing," he responded. Until now, VMware has been about server consolidation and running more applications on a single host. That's a savings in terms of the number of servers needed. The future of virtualization, however, lies in seizing the management advantages that virtualization offers. The future savings lie in reduced operating expenses, not capital expenses.

"In the long run, we know we have to do more than other vendors," said Maritz. With the low hanging fruit in hand from the early days of virtualization, it's time to think about how VMware will continue to grow its product line from a strong base. It's clearly thinking about capitalizing on virtualization's gains and move into managing applications in VMs in both data center and public cloud.

SpringSource started out as providing a simpler way to develop Java applications. Over the last two years, it's assembled a brain trust for deploying applications as well as building them. When it acquired Covalent, it got some of the best Apache Web server and Tomcat software expertise available. It's gained expertise since its early days in the crucial OSGi principles, which dictate how distributed Java objects can work together without a lot of prior knowledge about each other.

SpringSource added the Groovy and Grails scripting language framework to its line-up when Johnson recognized "there are some pretty compelling benefits to the enterprise" if it is added to the Java programmer's toolkit, he said at the time of acquisition. Then there was the addition of Web application management software from Hyperic.

All of this speaks to a future way of managing applications in the virtual machine and in the cloud. Being able to design applications for ease of assembly, management, modification and scaling out for high traffic performance could all be crucial elements of cloud computing. Is VMware crazy to buy an application development framework? Yes, it is, crazy like a fox.