Sun's Solaris success paves way for next-gen OS push
Sun Microsystems CEO Jonathan Schwartz loves to splatter the media with the line that Windows, Red Hat Linux and Solaris stand as the only operating systems of significance in the server kingdom. We've spent the last few years struggling to appreciate the seriousness of that claim. Sun's declining system sales failed to inspire much optimism about the company conquering the data centers of tomorrow with a
deflating “venerable” OS.
A number of recent items, however, coupled with a '06 spike in Sun's sales have tweaked our view of Schwartz's favored claim. It could well be that Solaris – of all things – provides the “iPod moment” Sun seeks.
Schwartz's justification for placing Solaris alongside Windows and Red Hat remains understandable enough. Sun has a gigantic customer base, immense ISV support and emphasizes its version of Unix far more than rivals such as IBM and HP. In addition, Sun spends a great deal of time courting developers and often includes Solaris angles in its various programs and applications aimed at coders.
For Solaris to “matter” in the sense that Schwartz uses the word, the OS must match Windows and Red Hat on forward looking terms – not a backward facing metric such as the installed base. Regrettably for Sun, Solaris suffers when analysts and customers make such “2010” comparisons.
For one, the Unix server market continues to shrink by most measures, while the Windows and Linux server markets grow at healthy rates. ISVs, customers and resellers all notice this trend and so divert their attention to the money making opportunities linked to Windows and Linux.
Linux enjoys an added bonus over both Solaris and Windows with its open reputation and backing from a wide variety of vendors. The open source OS benefits from healthy developer interest and diverse investment. Red Hat may dominate server-side Linux, but that does little to cramp the OS's communal cachet.
Schwartz, in particular, has tried to counter these issues facing Solaris, during his rise through Sun's ranks. He helped revitalize the x86 version of Solaris and then open-sourced the OS. Sun hoped that such maneuvers would break Solaris's ties to SPARC processors and create a developer frenzy of sorts with coders of all shapes and sizes hoping to explore the glories of the freshly released Solaris 10.
Where Sun bought into these hopes and dreams wholesale, we remained skeptical. Old Solaris x86 backers such as HP (Compaq, really) turned their back on the OS's Intel/AMD rebirth. Meanwhile, developer interest in OpenSolaris looked decent enough but few would confuse such contained enthusiasm with the industry-wide Linux obsession.
Through sheer persistence and substantial foresight, Sun may actually be eroding Solaris's “proprietary Unix” baggage, giving critics a reason to believe that the company has another turnaround in store.
For one, Sun's rivals have given up on knocking Solaris x86 in favor of, um, selling it. HP executives have recently been caught bragging about selling more x86 boxes running Solaris than Sun. HP also offers broad support for Solaris on x86 servers and even dishes out statements to customers celebrating this fact. Add IBM's tie-up with Sun around Solaris on x86 blades, and you have to give some credit to the Lazarus Edition of Sun's OS.
HP likes to bill the Solaris x86 push as a way of bringing disgruntled Solaris/SPARC types to its side. We figure that's just fine with Sun who would rather see a SPARC defector stay on Solaris than move to another OS running on x86 chips.
On both of these fronts, Solaris has gained that crucial freshness factor missing when Schwartz first started hammering away on the three OSes that matter campaign.
The Solaris revival becomes even more interesting as we look out to 2010, if our sources are to be believed.
For many months now, a crack team of Sun software engineers have beavered away on a secret project to create the “Solaris of the future.”
Refined details on this effort remain very slim, but we know that Sun's lead kernel engineer – and one of its top brains – Bryan Cantrill has been backing the project. In a recent blog post, Cantrill hinted that his team “is close talking about what we've been up to for the past year.”
Sun's CTO and R&D chief Greg Papadopoulos ordered the work on this software job, which, as we understand it, will result in an OS designed to power massive data centers. Yes, such operating systems perform this function today, but Sun's latest creation is said to be taking OS scaling to a new level.
Complementing this software work, Sun has started looking for ways to push Solaris deeper into the data center – namely the networking layer. Our sources have revealed that Sun hopes to woo Cisco as a long-term partner around this new flavor of Solaris.
Pipe dreams? Perhaps.
At the same time, it's hard to imagine any other company pulling off this type of OS planning.
Microsoft's release of Vista confirms that the company has no intention of innovating on the OS front anytime soon. God forbid, we depend on Red Hat and the open source army to craft a solid, usable OS of the future in a timely fashion. And neither IBM nor HP have demonstrated any interest in pushing their versions of Unix in new and interesting directions.
Plenty of start-ups crank away on file systems, specialized OSes and other virtualized applications that may do something similar to Sun. These companies, however, fail to have the reach or reputation needed to usher in something broad and spectacular in the server OS game.
Even Sun's staunchest critics – those who wonder how the company makes money off Java – would allow the vendor a unique position of prominence, if it can show broad, vibrant interest in Solaris.
The OS does not carry the razzle-dazzle of a new server based on a fresh multi-core chip. Nor do financial analysts translate software downloads into profits as easily as they do hardware shipments.
More savvy types, however, will see that Sun owes its resurgence to Solaris 10 as much as anything else. And a really dramatic turnaround – where Sun cranks up revenue by the billions – will only occur if Sun can maintain and then grow current interest in its venerable OS.
Luckily for Sun, such a Solaris revival now seems like a possibility. ®