Enterprise Unix Roundup: The Ghost of Unix Future
Here's what I wrote at the very end of the very last Enterprise Unix Roundup of 2006:
"In all, 2006 was a pretty steady year for Unix. It's still losing deployment ground to Linux and Windows, but the Unix vendors aren't even close to giving up. As long as Unix is tied to non-commodity boxes, there will always be a reason for vendors to try to keep Unix alive."
Apparently, there's a good reason I am not in change of a multi-million-dollar company. This whole notion of sticking with non-commodity boxes? Boy, what a crackpot idea that turned out to be, particularly if you're Sun Microsystems.
Of all of the enterprise Unix vendors, it's pretty much a slam-dunk to say Sun made the most dramatic moves in the Unix space in 2007. IBM and HP held their own, seemingly in a holding pattern of quiet deployments that kept them in the top revenue and units sold positions for the year. But, still trying to find a beachhead into the higher sales numbers, Sun put Solaris and its umbrella OpenSolaris Project through some serious paces.
Two major events dominated the Solaris scene. The first is Project Indiana, the project spearheaded by Ian Murdock of Debian GNU/Linux fame who joined Sun this spring. Indiana was Sun's effort to create a binary-based "distribution" from the OpenSolaris Project that IT shops around the world could download and evaluate. Indiana would provide a single point of download, as well as be a regularly updated version of OpenSolaris.
This was not without irking some of the OpenSolaris community members involved with other flavors of the OpenSolaris Project, such as BeleniX or Nexenta. Participants in those projects were none too thrilled that the final build to come out of Project Indiana would be named "OpenSolaris" effectively usurping the "official" moniker for Sun and marginalizing their own flavors.
By and large, the plan seems to be working, however. Reports of Solaris adoptions seem to be on the upswing. Anecdotally, throughout the industry I have heard more than a few reports of IT departments playing around with the OpenSolaris operating system. It will be interesting to see what the first-quarter 2008 numbers show.
The other big turnaround for Solaris was a actually a series of events where Solaris got picked as an OEM option on blade servers that Dell and IBM sell. Once Sun's mortal enemies, these hardware vendors put aside their differences and offered Solaris a place on their product lines. This pretty much blew a hole in my non-commodity box theory because clearly Sun is thinking that instead of selling purple boxes, it wants to position itself as a software- and services-oriented company.
This cross-platform play may be the best path for Sun. HP-UX is still optimized for the Itanium platform, and AIX is still geared toward POWER. If Solaris can get some traction on Intel-based commodity boxes, it could do well. Dell, without any legacy ties to a Unix, and as a company that's already shown a willingness to experiment with its new sales of desktop Linux boxes in the United States, may be the best fit for Solaris. Expect to see bygones be bygones and some stronger partnerships between Dell and Sun in the months to come.
Also casting a big shadow on all of the Unix vendors in the year to come will be the eventual release of Windows Server 2008, which has been pushed back to (Microsoft hopes) February 27, 2008. It's a common error to think that all Unix defectors go to Linux, when several studies indicate a majority migrate to Windows. If Windows 2008, already well past its original release date in late 2007, does not completely bomb like Vista has, then look to see the trend continuing. If, however, Microsoft's performance record holds true, then Unix outbound migrations might see a slowdown. Indeed, this might be what Sun is hoping to accomplish with Solaris catch people unhappy with the costs of their current flavor of Unix and get them into another Unix albeit a less-expensive flavor.
Meanwhile, the Linux vendors will keep going like great guns trying to move more into the enterprise market. Novell, aided and abetted by its new partner, Microsoft, will be making really big strides, at least until Red Hat figures out how to adjust to Novell and the revitalized Solaris. Canonical will be taking some baby steps into the enterprise space with Ubuntu Server, but I would not hold out for much from them for the next couple of years.
Of more interest will be how Apple plays its Leopard Server product into the enterprise. I talked to them two weeks ago and got some really interesting insights into where they want to be with Leopard Server. More on that next time, but I will say now that if Apple plays its cards right, its products could find a home in some key niches in the enterprise space as time goes on.
Last year, it seemed like everything was all about virtualization. Virtualization remains a big factor in the enterprise, I think the next interim step will be commoditization, as Unix vendors figure out new ways to get their operating systems onto a variety of platforms.
Until next year, peace on Earth and goodwill to all. And remember, only 30 years and one month to go until the Y2038 bug.
Brian Proffitt is managing editor of JupiterWeb's Linux/Open Source channel, which includes Linux Today, LinuxPlanet, and AllLinuxDevices.