Thursday, March 28, 2024

Virtually Speaking: IBM Goes High on Linux

Datamation content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

It’s not difficult to be nostalgic for LinuxWorld’s of yore. The show was once home to myriad announcements and filled with movers and shakers.

For East Coasters it was a not-to-be-missed event, and every winter the faithful made the pilgrimage to the Javits Center. Those on the West Coast got their chance to hobnob in August.

A few years ago, the East Coast show was dropped. Although it was replaced initially with a “summit,” then a small show in Boston, it too eventually disappeared.

Now, only the San Francisco show remains. And it’s no longer just about Linux, as the show is being held concurrently with the Next Generation Data Center conference at San Francisco’s Moscone Center.

That fact pretty much explains the spate of vendor interview requests between this week and last — the next-generation data center will indeed be a largely virtual endeavor, and it will require many tools, both physical and virtual to keep it humming along.

None of this is terribly surprising. Today, for many enterprises, Linux is just another operating system choice. In a clear case of be-careful-what-you-wish-for, it is as viable an option as some of the Unix heavyweights or Microsoft Windows, depending on the situation.

The other wrench is the possibility that virtualization’s gain is the operating system’s loss, although given the hypervisor’s potentially commodity status, it’s gain may well be negligible.

To shed a little more light on the rise of virtualization, IBM Tuesday released details about the behavior of its customers — more details, in fact, than it had ever released around these activities, Scott Handy, vice president of marketing and strategy for IBM Power Systems, told ServerWatch.

The story IBM tells is interesting. While most of the action in the virtualization space is focused around x86-based systems running Windows or Linux, IBM, which was into virtualization before virtualization was hip, is finding most of the action not far from where it started out — in high-end systems.

Approximately 21 percent of Power System servers (the former System i and System p product lines) that shipped last year were virtualized, Handy said. Contrast that with less than 2 percent of x86 systems. PowerVM (the virtualization environment for Power System servers) deployments tripled in the past year.

In the second quarter alone, 64 percent of POWER6-based Power Systems that shipped were virtualized with PowerVM, compared to just 21 percent in the same quarter last year.

As a result, IBM is bullish on the future of virtualization. “Virtualization,” Handy noted, is “the key glue,” enabling organizations to run up to 10 operating system images per core on any combination of Linux, AIX and System I.

Granted, the workloads these systems are running are far more complex than the bulk of virtualized x86 systems. But there are a lot them, and IBM has been pulling in market share in equal measures from Solaris, HP-UX and Linux workloads, Handy said.

The Linux side of the story is interesting. With workloads being continually brought on to Power in the form of Linux partitions on existing servers, Linux has moved far away from the edge, and “is sliding into the high end,” Handy said. However, it is “not being counted the way these things are counted,” as the boxes themselves are counted, not the virtual machines.

While its probably too soon to talk of overhauling the way servers are counted, as virtual machines become more widespread, at the very least, the way operating system penetration is measured will need to be re-examined. Assuming, the operating system continues to matter, of course.

Amy Newman is the managing editor of ServerWatch. She has been covering virtualization since 2001.

This article was first published on ServerWatch.com.

Subscribe to Data Insider

Learn the latest news and best practices about data science, big data analytics, artificial intelligence, data security, and more.

Similar articles

Get the Free Newsletter!

Subscribe to Data Insider for top news, trends & analysis

Latest Articles