Q&A: Red Hat's Jim Whitehurst Maps The Road To $5 Billion


Printer-friendly version Email this CRN article

The problem before, I think OpenStack kind of went through this hype-cycle and then kind of dipped down. Because it was open source, so many vendors wanted to jump on it and get there early, they were putting out alpha-level code. No vendor if they controlled that code would have put it out that early, but people were jumping forward on it.

I think now that's matured and there are only several companies offering an enterprise distribution, a lot of the others have winnowed down. We believe we were clearly the leader of that, and that's going to be a core component of infrastructure at our enterprise customers.

The interesting thing about it is that's kind of scale-out, stateless workloads. That’s also now dragging us in as a VMware replacement on stateful workloads because we had the same technology architectures in terms of the hypervisor -- KVM -- and RHEL; they're both components of our scale-up stateful workloads -- we call that RHEV [Red Hat Enterprise Virtualization], and our scale-out OpenStack.

A lot of people will say, "Hey, I'm a big VMware customer, but I want to talk to you about my scale-out, next-gen workload OpenStack stuff." And then they say, "Wow, why am I paying for VMware when I can use a similar management plane to look across both these things?"

So both sides of that business are growing because of OpenStack. And then where really CIO interest is -- I don't want to say peaking, but is exploding -- is around containers.

Is that specifically focused on Docker?

Yes. Docker is the dominant format and Docker itself is basically a specification for containers. The container itself is a Linux container, it’s a series of technologies in Linux that allow you to isolate an application and specify the amount of resources it can have, and guarantee amounts.

So we're one of the largest contributors to Docker, so we certainly support that format. We have several products that actually leverage Docker, including our PaaS, which is OpenShift, which is a Docker-Kubernetes combination for running and orchestrating applications and then a developer tool chain on top.

To go back to OpenStack, people have been saying for a while now that the challenge in working with that technology is finding the engineering talent. More than a year ago, you acquired eNovance, an OpenStack services provider, which I think was recognized as a really prescient move from a vendor to seize OpenStack talent. Is the dearth of talent still holding back the technology?

So we've now trained over 10,000 engineers. We have 10,000 Red Hat-certified OpenStack engineers out there. The technology matures, the level of engineering talent you need starts to drop. So at some point, it's matured enough and the number of people that have been trained, those lines kind of cross. And we're still not fully there, but we're getting much closer. That’s why you have some customers running pretty big installs of production applications. Also telco with NFV is kind of a killer use case that's really helping drive it forward.

Printer-friendly version Email this CRN article