Slashdot Mirror
Apple To Discontinue Xserve
Toe, The writes "Apple has announced that they are discontinuing their line of 1u rack-mount servers. With their usual understated style, the announcement comes in the form of a box on their website and a transition guide (PDF) to their low-end Mac mini server or their now-more-powerful-than-Xserve Mac Pro server. Attitudes about the Xserve have ranged from considering it a token nod to enterprise to an underpowered wimp to a tremendous value. Apparently, the migration to Intel processors removed some of the value of clustering Xserves, leaving them somewhat overpriced compared to other, more traditional offerings. The odd thing is that Apple clearly has shown they have the capacity for enterprise, but rarely the will to take it on. So, does the discontinuation of their rack-mount mean they have abandoned enterprise for their post-PC offerings, or are they simply acknowledging that their products aren't gaining traction in the data center? Or do they have something else up their sleeve for next year?"
It was the only way to look like a trendy douchebag in a datacentre setting.
OS X Lion Server will introduce the new "Lion's Share," and a new blade server appliance into which you can mount 9 Mac Mini's each with app store instant Lions Share server installs. Want AFP? Install Lion's Share AFP app on the mini. Want DNS? Install app store DNS app on another Mini! Roar! with Lion's Share!
Apple are dropping Xserve and Ubuntu is dropping X-Server
Your move Microsoft...
Summation 2
Xserve aside, OS X Server provides some very, very powerful tools. Many of them are based on open-source, but for the ~$1K price, a well-paid employee would be hard pressed to roll them all in less than $1K worth of time. And all these tools have no per-seat cost, unlike Microsoft solutions.
The question remains, of course, how seriously can people take OS X Server now that apple discontinued the Xserve?
OTOH, it makes a really nice home server, if it is a bit over-powered and pricey for that application.
> 1. You are stuck on one platform. It is like getting a Sun Solaris platform but worse because apple never really had a strong enterprise department.
They're Intel boxes. Run whatever OS you want on them.
Also, they're UNIX, so run whatever software you want on them.
> 2. You didn't get any real extra functionality over a Linux/BSD even Windows servers.
Setup times are far less time-consuming than Linux. Per-user cost is far less than Windows.
> 3. There is 0 fore-site on what will happen for the next version. What new features. Apple is too closed
Absolutely true, and a real deal killer in the enterprise.
>4. You had limited options. So that means you are paying for stuff you don't need
Somewhat true, but the Xserve is 1u. Most of the options are externalized.
>5. Limited server tools. Sure the Apple stuff is good but you need that one extra tool that apple doesn't support.
Then install it. The Xserve is UNIX. Also, most data centers have more than one machine, and hardly any have all the same brand throughout.
>Like Apple or Hate Apple, it really isn't a good server platform.
Well, they still make servers, just not rack-mounted ones.
People hear *nix and, if they think anything at all, they think "server."
Or they think "I know this!" and then check quickly for Velociraptors.
which is totally what she said
The Xserve has been largely redundant since Apple discontinued the Xraid. When you pair them up they make great file servers, the publishing company I used to work for loved them (yup, that's right, there *are* people for whom Apple servers make sense, go home haters).
Seeing as how there's nothing you can do with an Xserve that you can't do with a Mac Pro, the only difference is the rackmounting. Considering the way forward is Xsan, that's completely optional now even if all your storage is rackmounted. The SAN controller can be on the other side of the building as long as your fibre reaches it.
Nice as it was, goodbye redundant product. You'll be missed, but not for long.
I think you are on the right track, but you have cause and effect backwards. Apple strives to be a consumer company. I was confused by Apple's Xraid/Xsan and Xserve products, because they don't really fit in the same milieu. If anything, I wonder if offering Xserves and Xraids was just a way for them to kill harbored distrust of OS X inherited from OS 9. After all, OS X had some big hurdles to overcome from OS 9. Supplying even a couple universities with Xserves demonstrated that OS X and Apple in particular were making high-performance machines, a worthy continuation of their NeXT legacy, and dispelled any fears about inherited OS 9isms. So from this standpoint, the product line was a success, but it is no longer required.
From another standpoint, remember that Xserves were first brought onto the market was during the bubble, before "the cloud" was a thing. My first employer had an Xserve simply because he found the idea of managing it better than the idea of managing a hosted Linux server. Colocation was cheaper than paying for a managed server. For small business owners--particularly Mac software developers--it made more sense to them than learning how to administer Linux or paying another employee to do it. Familiarity is worth something.
Remember also the market conditions when Xserves were brought out. They weren't the only vendor selling their own weird Unix in a 1U. SGI, Sun, and HP at least were also selling their own servers running their own Unixes. The market was nowhere near as homogenized as it is now, and it was plausible at the time that OS X Server could become as important as the competition. It turns out people don't buy servers for the same reasons they buy desktops. That wasn't necessarily obvious five or ten years ago.
Apple's hardware support was abysmal for their servers. And by support, I'm not talking about drivers, I'm talking about their ability to fix a broken system.
I've called Apple to get parts for failed Xserves, and they have taken WEEKS to ship for systems covered under applecare. They also think it's entirely fine to tell you to bring an Xserve in to an authorized repair center. I mean, *what?*
Just because a server is available in a 1U form factor doesn't mean it's an enterprise system. You can't support enterprise hardware the same way you support iPods.
It's not the 'ooo shiny' products that everyone knows Apple for, but their Server OS is quite good. As are some of their clustering capabilities.
After trying to setup a linux cluster, XGrid is nothing short of Magic. It's a check box in a system control panel. You can let anyone use a computer or password protect it. Buy 1+ Macs. Check "Allow for use on XGrid" (and even set to only use when it's been idle). Anytime you compile something in XCode, all other available Macs will be used. No setting up which servers to use in a .distcc file.
Their Server OS is also pretty polished. I know the hard core command line junkies think that everything should be done with vi/emacs and only configured from there. But not everyone wanting to run a server has that expertise. If I had to suggest a server to a friend for a small home business. I'd suggest the MacMini Server. Mail, Web, Jabber, OS Updates, Time Machine, etc.
I suggest checking it out (not sure if they have the server OS setup in any Apple Store) before knocking it.
If you worked in an organization where you hand to manage hundreds of desktop Macs it would dawn on you. I know places like that are not exactly everywhere, but they do exist...I know because I work at one. After our current Xserves reach the end of their useful life, there will be absolutely no enterprise class hardware that can be used for management and deployment of desktops. I'm sorry, but we will not be putting Minis on rack shelves for services (Open Directory, for example) that need to be available 24x7. The lack of redundant power supplies, lights out management and hot swappable drives is just not acceptable in an enterprise datacenter. I also cannot give up 6u per Mac Pro just to get the same 8-core speed that is currently available in a 1u box while I'm giving up those availability features. It's just not going to happen.
We're already pondering what life is going to be like with Windows-only desktops. It's possible Apple is going to pull off some announcement about OSX server availability on VMWare/HyperV or something (which would be acceptable for larger environments I guess) but I can't plan ahead for what they *might* announce. Frankly, it doesn't give me confidence in the future of OSX server in general. Without large deployments, people won't be needing things like Open Directory, so that could easily be dropped at some point, for example.
They are effectively removing themselves from consideration in large environments. Just a week or two ago, there were rumors they were going to make a bigger play in the enterprise space and I was anxious to see what they had in the pipeline. Now, suddenly we're looking at abandoning OSX as a platform almost completely at hundreds of desktops. We were about to put in an order for 50+ iMacs for the second or third time this year alone, but now those plans are on hold until we can get a new long term plan together. As a result, it is definitely going to be a factor in future iPad/iPhone deployment (which has been pretty positive thus far).
I never doubted Apple was evil, I didn't know they were dumb too.
teeker
Not at all, it means that the market for companies that have racks and want to run OS X Server is small. Now they have the Mac Mini Server, they have a product that can go into small offices that don't have somewhere for a rack-mounted system.
The XServe was never a product that Apple created to sell. They created it because they have a lot of data centres of their own (to drive their site, the QuickTime Movie Trailers hosting, Apple Store, iTunes Store, and so on) and they didn't want to be buying a load of servers from a competitor to run them. They sold it because, having already designed and built it for un-house use there was no reason not to, but the potential market for a rack-mounted OS X box was small enough that it wouldn't have been worth their while designing it just for sale.
So what does this announcement actually mean? That they are no longer planning on using XServes in their own data centres. My guess is that they're planning on having their ARM team design a Cortex A15 SoC with ethernet, crypto hardware, and SATA on die and make tiny blade servers for internal use. They almost certainly won't want to ship OS X Server for ARM for external use, because supporting another architecture would be a lot of effort for little return, but they might do if the market looks big enough.
I am TheRaven on Soylent News
Yes, they have something else up their sleeve. Did anybody notice Apple's "enterprise services agreement" with Unisys? http://www.eweek.com/c/a/Mobile-and-Wireless/Apple-Unisys-Agree-to-Enterprise-Services-Deal-Report-788654/ Did anybody notice the 54% drop in Unisys's profits, along with a drop in server sales? http://finance.yahoo.com/news/Unisys-3Q-profit-sinks-54-pct-apf-3818156357.html?x=0&.v=1 So, Unisys is an enterprise computing company looking for a way to save its server business. Apple is consumer electronics company with enterprise ambitions, enterprise software, but no enterprise distribution network. Apple just announced it is dropping its server hardware line, a little over a week after announcing the deal with Unisys. I know it is fashionable to dismiss Apple's enterprise computing ambitions. I was at an Apple Developer's seminar a couple years ago where they were showing off the new version (then) of MacOS X Server. The entire focus of that seminar was on how Apple was adding features to MacOS X Server (and even licensing things from Microsoft) to make OS X Server more suitable for the enterprise. I predict Unisys will start offering MacOS X Server on Unisys server hardware soon. Apple may even end up buying Unisys.
Actually, I think the previous post is probably right, at least in the long run.
If there's one thing the popularity of iPhone/iPad had demonstrated, it's that most people don't really use their computers much. They have a hugely capable desktop machine that they use for "facebook", email, and "youtube", and that's about it for most of them.
I'm a pretty hardcore penguinista myself, but even I doubt that a standard full-service (by today's "PC" standards) Linux desktop will ever conquer the market, or even a large minority of it. However, I think the current "desktop" market is mostly doomed outside of "enterprise" and hardcore power-user settings. Now that "consumer" gadgets have gotten cheap and powerful enough to do what the great majority of "users" seem to do with their computers, there's no need for it any more. All those people who are "completely befuddled when they don't see the start button" will be migrating their way over to even-simpler environments like Android and iOS and perhaps Windows 7 Series 7 Phone 7 Series (or whatever they were calling it), which I actually kind of expect will cannibalize BlackBerry for corporate users.
My personal prediction: Microsoft is busy fossilizing into the New IBM (firmly embedded in many "corporate" environments but fading out of the "consumer" market), while Apple clamps down on its users and gets increasingly ruthless with its market control to become the New Microsoft. I expect Linux to grow solidly on the internet server side and on corporate servers.
I actually expect the Android/Apple landscape in the "consumer" side to end up looking a lot like the Microsoft/Apple market now - I'm guessing we'll end up with a solid majority made up of various Android devices, with Apple being a minority (but a relatively large and reliable one).
There, a free wild prediction, and you didn't even have to look at ads on ZDNet or some other commercial publication to get it.
tl;dr: Yes, I agree that Microsoft will likely hold onto the "traditional desktop" market for as long as that market stays around, but I don't think that market is going to exist for that much longer now.
Hacker Public Radio is our Friend
That's a whole lot of speculation.
The way I remember it, Apple sold a lot of Xserves into media environments, for digital video processing, basic file storage, etc. Musicians and A/V professionals have a natural affinity for Macs and little interest in maintaining servers, so a plug-and-play server that worked with their Macs was a natural choice. Unfortunately, it's not a particularly large market.
Xserves were nice machines, but building and maintaining bulletproof server hardware -- including continually producing new models that keep up with the ongoing upgrade cycles from Intel and other component vendors -- just doesn't make sense if the products aren't competitive in the market. And Apple's servers weren't going to be competitive until it started shipping models with Linux and/or Windows Server as an option. Instead, Apple tried to be Sun and found out it simply didn't have the expertise and market savvy to be Sun -- and then, look what happened to Sun.
They almost certainly won't want to ship OS X Server for ARM for external use, because supporting another architecture would be a lot of effort for little return, but they might do if the market looks big enough.
So they're going to use it exclusively in-house, to the extent that they're going to replace all their Xserves, but they don't have enough faith in the ARM port to sell it? Just the fact that they put it into production use in-house means they'd have to "support" it. I think you're reaching.
Breakfast served all day!