NYT Ponders the Future of Solaris In a Linux/Windows World 340
JerkBoB links to a story at the New York Times about the
future prospects of Sun's Solaris, excerpting: "Linux is enjoying growth, with a contingent of devotees too large to be called a cult following at this point. Solaris, meanwhile, has thrived as a longstanding, primary Unix platform geared to enterprises. But with Linux the object of all the buzz in the industry, can Sun's rival Solaris Unix OS hang on, or is it destined to be displaced by Linux altogether?"
solaris is the new AIX (Score:5, Interesting)
it'll just be a niche product.
personally i think it's sad sun blew their chances with solaris, it's superior to linux in security and performace.
Solaris 10 (Score:5, Interesting)
If your only experience with Solaris is v8 or v9, you really need to check out Solaris 10. It is a complete night and day difference in ease of use and features. Add to that the volume of useful enterprise management software from Sun (the N1 stack, and now the new xVM stack) and you have an enterprise that is a dream to maintain.
I've been doing straight Solaris 10 admin for the last 2 years (linux for 4 years before that), and shortly will once again be taking a position that will be 99% linux. I will miss Solaris 10. I still love both OS's, but Solaris wins in my book at the moment.
Re:Performance (Score:3, Interesting)
ZFS is really not that impressive. Only if you are building systems with giant disk subsystems - which is exactly where you will be buying SAN at this point instead. We just bought a HP 5000 series SAN with 138 disk drives in it... does self-healing and provides up to 1.4TB/s throughput, doesn't use ZFS. ZOMG!
It's the hardware (Score:2, Interesting)
Re:How about some technical analysis (Score:2, Interesting)
With all the GCC bugs Linux has? With the poor track record on security?
Linux might be fine and dandy for web monkeys, or huge data crunching. It's cheap. It's fast. But that's not all there is to it, at least for some stuff.
For apps, on Unix, thanks, but no thanks. Not with that sloppy "release early, release often" process. Even the LHC project had security problems due to Linux. Wake up!
Re:Not all the best features are technical (Score:5, Interesting)
Please elaborate with what aeras and tasks that Linux is so good at that Solaris
can't touch?
I use both Solaris and Linux in my environments and Solaris 10 is by far the superior
OS in my opinion. We have Solaris servers on both SPARC and AMD64 and Linux on AMD and Intel 64 bit hardware.
We had migrated a number of Sybase instances to Linux, but we kept having reliability and performance problems, so we migrated them back to Solaris but Solaris 10 on AMD64 boxes
and we've been extremely happy with the results.
Our company is current migrating all of our market data servers to Solaris AMD64 servers in Zones and will reduce the number of Linux servers which stand at 25 to 4 X4600s running Solaris 10. In our testing of Solaris on x4600 as opposed to DL585s (same CPU and memory configurations) we have seen a large performance gain and cheaper operating costs, since we don't have to pony up a RedHat license for each server.
Re:Virtualization makes Solaris less relevant (Score:3, Interesting)
Re:Virtualization makes Solaris less relevant (Score:3, Interesting)
It's kinda funny watching things go in circles. First it was hardware built to run one specific app. Someone said lets build something so we can run multiple apps on general purpose hardware. So the Operating system was born. Now we want to run multiple operating systems on even more general purpose hardware.
More like it's the support (Score:3, Interesting)
I've worked in Sun Shops before, and I've seen Sun support folks come in to repair 15 year old boxes that were running mission critical databases. Also, if you write sun certified software, they tend to bend over backwards to ensure it will be backwards compatible. I've even seen Sun send engineers when a Solaris 6 App stopped working in Solaris 8 to help the shop solve the problem.
That may not seem like much to you, but if your a decent sized business that is making millions of dollars per year and it has to work, Sun is a worthy look if for no other reason than you only have to develop that application once with reasonable assurance that it will work on future versions of the OS.
Re:Or else... (Score:5, Interesting)
I use the Solaris 10 JDS everyday at work and also run Ubuntu and Solaris 10 x86 at home, with zones on it. Basically the "apt-get" you're looking for is called "pkg-get" and is available from blastwave.org.
The future of Solaris on the desktop is not as exciting as that of Ubuntu, or any other wildly popular Linux distro. The enterprise future of Solaris is way more exciting IMO. The reason is this; Solaris 10 Zones are ready for primetime enterprise whereas Linux is still being pondered and in most cases not being taken seriously due to the open source nature of the beast and the sheer number of different distros, many of which lack and enterprise level support. We had a Red Hat box in our DC and we retired it. Meanwhile we're approaching 400+ Solaris 10 zones and we're coming in at *HALF* the price of the VMware solution that the Windows side of the house provides. Guess who's growing faster? Even with more expensive boxen our solution is a better value and provides a very solid framework for many, many environments.
As an OSS advocate and 20+ year SunOS/Solaris admin I will welcome both operating systems in my home and data center. Like the man said above, you know one, you practically know the other. It's all *nix in the end.
Re:Nahh.. its not going anywhere. (Score:3, Interesting)
... except Sun appears to be too busy turning it into a desktop OS. Many of the underlying subsystems that have been added in the past few major releases need some work to make them enterprise ready. [My favorite is the lack of group and netgroup support in RBAC. How can they possibly push it as a sudo replacement without that?] Instead, they are too busy screwing around with things like WiFi support and replacing Jumpstart with a system that doesn't even support begin/pre and finish/post scripts... and asking the community to justify it. WTH?
Sun is about to lose the enterprise completely and they don't even know it. ZFS is interesting in a 'last decade' sort of way, for those machines where you care about the local storage. Lustre--which only works on Linux--is what is really needed to take on this decade's problems. While they are porting it to Solaris, why should I as a large enterprise use it on Solaris when I can use it on Linux today?
There are lots of things Sun got right with Solaris. But it is clear that they simply couldn't (or, more likely, wouldn't, given the arrogance of your typical Sun engineer) execute and keep up with what was happening around them. Too much time trying to build 'perfect' solutions instead of giving us, the customer, what we wanted even if it meant the interface wasn't Stable.
Re:Not all the best features are technical (Score:1, Interesting)
I wouldn't leave Sun Sparc arch out to die just yet.
The new Rock processor due out next year sounds pretty dam cool! Seems (to me) they are just having demand volume/price ratio problems.
Also when Solaris/open solaris is better tuned to run on x64/power archs you never know what might happen.
An operating system that is a always evolving at the rate SunOs is won't die anytime soon, not at least when every other OS is gutting its open source features.
Re:Or else... (Score:3, Interesting)
Even with more expensive boxen
Which is where the discussion ends with me. I use Linux because it cuts my costs for fielding compute power. Sure, I could use Open Solaris, but why would I use code away from the center of the action? That's where the bug fixing is happening. The key thing for me is cheap hardware and cheap software. I use a heavily replicated environment, so one box going down isn't a big issue. Cost of the farm is. So Linux on cheap x64_86 is where it's at for me.
Re:Not all the best features are technical (Score:3, Interesting)
since we don't have to pony up a RedHat license for each server.
Um, is this an 'am I stupid' test? If you have any sort of test environment, you run redhat on the test server and CentOS on production. Personally, I've done fine for four years now with Fedora in production. No support dollars. The precise reason you run linux is so you can multiply your servers without marginal SW cost. If you need to run Oracle or something that you can't multiply, then Solaris on big Sparc makes sense.
Re:Or else... (Score:3, Interesting)
My brother in law who works at Intel says they no longer use Solaris for CAD/Simulation - its all been moved to Linux.
Also have a friend at Cisco who says the only Solaris machines they have in their department run FrameMaker and that if Adobe ever made a Linux version they'd abandon Solaris entirely.
Re:PC-BSD anyone? (Score:3, Interesting)
And how about the other way. For me, the big reason for switching from Linux to FreeBSD was something very simple (although there ended up being a load of reasons for staying): Sound just worked.
This was back in the FreeBSD 4.x days. I had a cheap on-board sound card with no hardware mixing. On Linux, I got one /dev/dsp, and so only one program could play sound. If I ran KDE programs, they would all talk to the KDE sound daemon, and so they could all play sound. Same situation in GNOME. Only, I wanted my (KDE) Jabber client to be able to notify me of messages, my (GNOME) email client to be able to notify me, and XMMS to be able to play music in the background. XMMS didn't support either sound daemon at the time, so Linux failed.
FreeBSD, on the other hand, had a sysctl to set the number of virtual devices. I then had /dev/dsp.0 to /dev/dsp.4 - one for GNOME, one for KDE, one for XMMS and one for BZFlag. Configuring each of these to use the correct device was mildly painful, but it did work. Then came FreeBSD 5, and suddenly all of this configuration was obsolete - each program that tried to open /dev/dsp would get a new channel. I set the maximum to 16 on boot, don't bother configuring sound daemons, and laugh at all of the issues that Linux users are having with PulseAudio.
A kernel is supposed to do two things well:
This is an example of Linux failing massively at the second of these - if it's possible to multiplex a device, it should be multiplexed by the operating system. If you want to do it in userspace, fine, but support standard APIs for doing it. OSS is the standard sound API for all *NIX systems with the exception of OS X, and it makes playing sound very easy (four lines of code - one to open the device, two to set the sample format and rate, one to start writing the audio). A userspace sound daemon could implement this with something like portalfs, but none do, they all introduce their own APIs (and most have terrible latency).
Linux also tends to fail at the first of these too. Each supported architecture populates /proc slightly differently, so you need to code around these differences or, more commonly, use an existing library where someone else has already done it. Rather than writing three lines of code issuing a sysctl call, you will link against several hundred KBs of someone else's library. Which do you think is more maintainable (and auditable)?
Re:Or else... (Score:3, Interesting)
Been done already. Life is passing you by.
OTOH, all of the hype surrounding "big iron" is
compelling up until the point where you have to
juggle conflicting SLAs, deal with zones that
are underpowered or you manage to find that one
point of failure in one of those "big iron"
machines.
Then you get to see 10 production systems go
down rather than just one.