Solaris x86 to be available on IBM Opteron and Xeon based servers. The Register has more details.
My high level non factual 5 second dismissive take on this: Solaris is good and people still like it, for now. Sun hardware is fantastic, but way too damn expensive. IBM can sell crap to anyone, and is still somehow respected in the "Enterprise." The marriage makes sense, IBM backing an OS that some people want (and still offering other OSes like SuSE Linux too).
And now for some ranting:
In the long term, Solaris x86 is just a stop gap. Linux will "win" despite efforts to beat it. Linux is entrenched *because* it was the first major OS (along with OpenBSD and a few others) to recognize the power of open source. The GPL and Linux are pretty much unstoppable, despite years of people whining about the "cancer" and simultaneously claiming it will never amount to anything.
OS server software will be Linux on x86, lots of boxes, cheap clusters, a commodity. Enterprise shops will still want to buy support, and tested distros, etc, but those are available already from Red Hat, Novell, and so on. We are in the "last throes" of the non Linux data center, really.
Solaris may be better than Linux in many ways, better threading, better memory management, better desktop, better whatever, and none of that matters. I don't need the best, I need the most readily available and adaptable open solution that I can cluster to scale (and Linux is no slouch in any of those areas for the record, the point is that stuff is not the most important factor).
Solaris "x86" itself is an admission of the shifting paradigm. 10 years ago, when many people were already using Linux (myself included), Sun scoffed at x86 altogether - much less open source. And when you install Solaris, even if on Sun hardware, what do you have to do to actually make it useful? Install Sendmail, BIND, Apache httpd, the GNU Freeware tools, and so on.
As more and more people realize that is just silly, and use Linux to begin with, Solaris x86 will be even more marginalized. IBM will sell some in the meantime, but in the long run Solaris can't "beat" Linux.
And don't get me started on other "server" alternative OSes such as Windows DataCenter. That just won't scale economically, support wise, security wise, nothing. I know many people use it now, and it is a capable product, but it is not going to have any affect at smart companies that recognize the need to commoditize and scale (and throw virtualization into that mix too). And, in the long run the list of organizations coming to that realization will only increase, because it is truth, not because of any advertising or marketing or anything else. You can't fight the power, simple, whether or not you are "better."
Comments
How can I put this....
You're wrong. Windows server, HP/UX, even AIX and Solaris will be around in the big boys' data centers for a long, long time to come. It's just reality. Free beer, (and don't even get me started on all that bullshit about free speech being the reason of linux's mass popularity, (what there is of it),) will only get you so far...
So if you haven't already flamed me back, read on and I'll tell you why.
M
O
N
E
Y
It is a bitch to make money off of something that others are giving away. Who makes the decisions at the big companies with big data centers? IT staff? Administrators. Hah! It is the CxO who gets smoozed the best by someone. So who's gonna have more cash to throw at this, an OS that sells for a gazillion a proc or someone who is trying to scrape into the Linux support/sales market? Please. Your faithfulness and enthusiasm are endearing, but big biz rules, and big biz likes to spend money on people who spend money on them.
I just disagree.
I just disagree.
Big biz does rule, which yes means money, and that is exactly WHY Linux and the "commodity" will prevail in the long run.
You don't consider Google/Amazon/Yahoo/eBay etc to be big business? Other companies are just behind the curve because they are, frankly, not as smart - or, not in as competitive a setting for their primary business. They will follow rather than lead, but they will show up.
Once you have to COMPETE with Google/Amazon, or you have to compete with a company in your industry kicking your ass on server costs/security/reliability that will start to matter a lot more than who your CIO plays golf with and so on. Unless of course hardware/software and bandwidth costs don't matter at all to your business. Though, I would argue once you need a data center, HP-UX/AIX/Solaris, those things do matter.
"It is a bitch to make money off of something that others are giving away. "
That's a common sentiment, but not all that accurate. There are a ton of companies making money in the services and support side of open source products. That market will actually GROW as the data center invariably and inevitably migrates to commodity boxes.
(And it's possible Linux might not be the commodity, it has a hell of a head start, but Linux itself is not the point, the commodity nature is.)
Money Does rule... But the .com days are over.
Currently a Solaris admin supporting 200+ box's. Yes, we still have some new projects coming in that require Solaris/SPARC, but there are a more projects coming along that want to use the "cheaper" Linux/x86 solutions. Not to mention Oracle RAC plays nicer on Linux box's than it does Solaris.
Solaris/x86 is too little, too late and a duplicate effort.
Not to mention Sun support has/is going down hill. Most of the time lately I know more about their software/equipment then their techs. All the good ones I know that use to work for them got better paying jobs w/ 3rd party support companies.
In regards to money, We have none. Our customers have none. The last 2 big projects where some "shady" things happened where lots of money was... Well... Lets just say those that where part of those deadend projects are gone and the software/hardware shelved or repurposed and we went back to the "free" solutions that worked.