The rPath technology first takes an application and analyzes its code to determine how it interacts with other pieces of software, Adam says. The result is a “complete bill of materials” that shows what components are essential, and how the application will be affected by changes to the surrounding software. This way, Adam says, each application can be separate and self-contained without introducing confusion.
Adam notes that this increases efficiency by reducing maintenance needs. For example, he says, in the past two years, there have been 200 security patches for the basic Linux system that rPath itself uses. But because rPath applications are self-contained and each application is only tied to certain parts of this operating system, only 6 of those 200 patches, or 3 percent, had to be installed.
“Reducing the application’s footprint reduces the number of things that can break, and the number of things that can be broken into,” Sorofman says.
If an application does break after a change, Adam says that the rPath platform makes it easy to restore the application to a previous version. A similar ability is offered by the automated debugging tools commonly used by programmers. “Every professional software developer on the planet takes this capability for granted, but it hasn’t been available for IT operations staff before,” Adam says.
The rPath technology serves as “a bridge between the old and the new,” says Reuven Cohen, the founder of the Cloud Computing Interoperability Forum, a group that works to facilitate standards for cloud computing. Cohen, who is also the founder of a Toronto-based cloud-computing company called Enomaly, says that rPath is one of the first companies to attempt to simplify application deployment in this way. However, he says that the company may have some marketing battles ahead: “It’s an application that you don’t realize you need until after you need it.”