Red Hat's critics say containers are the wrong idea, and too focused on the operating system. I say maybe Red Hat knows a thing or two about pleasing developers.
Critics of Red Hat's newfound emphasis on Linux containers say it's a way for the company to keep its operating system at center stage. Developers won't pick up on it, they predict, because developers do things in their own interests, not Red Hat's. The company thus far has failed to convince many of them they should buy into its container approach.
That Red Hat has adopted an operating system-centric approach, I agree fully. Proposing containers as a better way to package cloud workloads sidesteps the issue of whether you need to convert to Red Hat virtualization. Many of Red Hat's customers were sidestepping that issue as well. They had already adopted VMware and were happy with it. But if these same customers could be converted into container users, then Red Hat Enterprise Linux (RHEL) would be required for future operations.
Focusing on containers changes the terms of discussion from How should we do virtualization? to What's the best orchestration of a workload for the cloud? and How should the operating system be provided for that workload? Containers offer both convenience and a slimmed-down set of answers to those questions, so it might be said Red Hat's adoption of containers isn't completely witless. Through containers it's bidding for its own future role in the cloud.
In some quarters it's become a trend to say the operating system doesn't matter anymore. That was a sore point with Paul Cormier, Red Hat's former VP of engineering, now president of products and technologies, as he spoke at the Red Hat Summit April 15. "The traditional operating system has all but disappeared, according to the former CEO of a leading virtualization company," he noted. He was referring to Paul Maritz, former CEO of VMware, now CEO of VMware spin-off Pivotal.
Red Hat doesn't want the operating system to disappear. It wants to keep it uppermost in customers' minds. Its view of Linux containers as the future packaging for cloud workloads is one way of doing so. Containers require awareness of the Linux being used in the container and a compatible Linux kernel on the cloud host. If ease of operation is aided by the fact both are RHEL, so much the better, from Red Hat's point of view.
That's why there was controversy in March as Rackspace and IBM lined up behind the new foundation that will govern Pivotal's Cloud Foundry open source project. They're betting that backing containers is picking the wrong horse. The virtual machine is too deeply entrenched and tied into overall datacenter management to be dislodged. They have a point.
But Red Hat is damning the torpedoes. Captain Cormier said the good ship Red Hat will redouble efforts to incorporate containers into Red Hat's platform-as-a-service, OpenShift. It will also contribute to Project Solum in OpenStack, its way of injecting containers into the OpenStack cloud. How widespread will be the backing of that project, if it's viewed as primarily about keeping Red Hat's operating system front and center in the cloud? The backing will be embarrassingly narrow and perhaps confined to Red Hat and partner Docker, if that's the case.
Developers have yet to be heard from in this debate. There are always good reasons for them to stick with what's known and with the status quo, but both they and cloud-service providers are always looking for a faster, cheaper, and better way to do things.
Brian Stevens, Red Hat CTO, took the Red Hat Summit stage April 16 to say Docker has 350 contributors to its project, not all of them Red Hat's. About 40,000 developers have been trained to use Docker, not all of them
Charles Babcock is an editor-at-large for InformationWeek, having joined the publication in 2003. He is the former editor-in-chief of Digital News, former software editor of Computerworld and former technology editor of Interactive Week. He is a graduate of Syracuse ... View Full Bio
Google in the Enterprise SurveyThere's no doubt Google has made headway into businesses: Just 28 percent discourage or ban use of its productivity products, and 69 percent cite Google Apps' good or excellent mobility. But progress could still stall: 59 percent of nonusers distrust the security of Google's cloud. Its data privacy is an open question, and 37 percent worry about integration.