Why Apple axed Xserve, and how it can reenter the sever market

1356

Comments

  • Reply 41 of 101
    Quote:
    Originally Posted by irv999 View Post


    The management ad deployment breaks into a few categories.

    1. File services

    2. and management



    File services for the avg small business, the mac mini is fine but it will not serve the needs of the other part of this market.



    1. Video production (Which you mentioned)

    2. LArge file IO design / print..

    iSCSI doesn't work because its throughput won't work for either application. Even if you decide that a mac pro is a good idea, everyone's infrastructure is rack based systems. 19" wide.. the mac pro because if its incompatible size and monitoring ability cant be hosted anywhere at a reasonable cost (19 U is expensive to lease), it can't be monitored.. IT is a terrible idea.



    You need a server to fit inside a rack and be monitor able. The mac mini works for the rack system sorta, but has no mass storage, and the mac pro just fits neither application.





    This is one of the reasons why the Apple store is a bad idea. Some guy walks in and says "I have a business and I need a server". before they would send them to a xserve.. IT was server class, it had a warranty which was better than most, and fit inside a rack.. Apple came out with a mac mini, and now the apple store says .. "buy this it is perfect". The turned a 4K sale into a 1K sale..



    So what does one do?



    If you are a design firm, and you have large files.. you buy a windows server with extreme ZIP. IF you are a video client and you use XsAN... you are in TROUBLE! Good luck. Take you rack which used to hold 48 servers and now it holds 6.. AWESOME!.. And if your master controller for XSAN breaks.. cross your fingers and hope someone comes down to fix your machine.. because there are no user installable parts OR guarantee on part deliver time...



    apple this was a dumb move..



    I'd argue that for XSAN its not as big of a problem as you mention. XSAN is an apple branded version of the Stornext file system. Stornext has clients for a multitude of platforms. The Mac Pros that editors use can still run the Apple client, and the MDCs can be migrated to Linux when the Xserve hardware finally gives out. The lower cost of a stripped down 1U from Dell or HP would probably offset the increased cost of a Stornext license key for Linux. (Apple's charged less for XSAN keys than is typically charged on other platforms).
  • Reply 42 of 101
    Quote:
    Originally Posted by Market_Player View Post


    I am running OS X Server inside Oracle VM VirtualBox on a Dell Power Edge as you are reading this.



    http://dlc.sun.com/virtualbox/vboxdownload.html

    http://www.virtualbox.org/wiki/Guest_OSes



    Enjoy



    Shame on you, that not allowed according to the OS X Server licensing terms - i.e. illegal.
  • Reply 43 of 101
    mactelmactel Posts: 1,275member
    If Apple OEM'd OSX Server edition it would not cost a mere $500. Server support is expensive. Apple would quickly up that to be more in-line with MS.
  • Reply 44 of 101
    asciiascii Posts: 5,936member
    I don't think there's any money to be made in server software. Remember when people thought open source was going to take over the world? Well in the server space they were kind of right. The kind of apps that open source produces -- highly configurable, no documentation, designed to be used by geeks, very basic GUI or command line only -- are very suited to this space.



    Hardware vendors of course can make money through support, but software vendors will just be in a fight to the bottom with OSS. Apple are financially sensible to get out of this market.
  • Reply 45 of 101
    What have Apple got to lose by discontinuing the Xserve? As small as 10,000 units is, it's 10,000 units. They would have announced Xsan's demise and probably Final Cut Server at the same time if they were abandoning the data centre completely.



    Mac OS X Server certified HP server configurations will be announced some time before January 31.
  • Reply 46 of 101
    Quote:

    Apple does however have the capability to relax its Mac OS X Server licensing to allow partners to install it on third party hardware, which would be a bigger win at lower costs for the company than continuing to develop unique hardware at a loss.





    Licensing is the way forward. Take a page from Microsoft's recipe for success. License software and make $$$.





  • Reply 47 of 101
    This article was one of the wackiest I have skimmed over.



    I concur with filburt's posts on page 1; he has his head screwed on.

    Quote:

    Aside from pretty hardware and Mac OS X Server's configuration tool, there's nothing particularly unique about Xserve. A competent admin can put together a Linux rack with necessary tools for significantly less.



    Quote:

    Mac OS X Server's features are based on free open source apps that runs on Linux (and installed on many Linux server distros, minus pretty admin UI).



    Outside of Final Cut Server type software, there is literally no reason to go with OS X Server, and countless reasons not to go with it. Cost, performance, speed of patches and security updates, the fact that virtually all the software is FLOSS (open source, runs on Linux, BSD, Windows as well), etc etc.



    $3000 starting price for a $1000 PC is Apple's 'new' server. Nuff said. Apple's server hardware has failed because there is quite literally no reason to buy it.



    Apple is better off focusing on gadgets and consumer-space servers for media.
  • Reply 48 of 101
    Quote:
    Originally Posted by Dick Applebaum View Post


    .



    IMO, Apple needs a Home Server Solution with RAID, local TimeMachine backup, co-ordinated with cloud backup.



    Something to Store iTunes and iLife content and serve it tp AppleTV -- ultimately to be replaced with modstly cloud stouage of your content and the home server being used as fast-access staged storage from the cloud.



    .



    I have been looking at a DataRobotics DROBO box to hold my photos, music and other critical files securely, but I would prefer the Apple solution you are proposing.
  • Reply 49 of 101
    Quote:
    Originally Posted by Daniel001 View Post


    Interesting article, although Apple have probably already determined their server strategy, and it's probably not this.



    Absolutely agree. Apple will not sell OS X server as a software only package and will not partner up with other hardware vendor. The logistics for a software licensing are not in place and the overhead and restructuring that a move like that would take will eat up the supposed $500 licensing fees. Plus a licence for OS X Server is $499 because it is actually an upgrade license.



    What is a lot more likely is that Apple is rapidly moving into the Cloud space and that NC data centre is about to become the largest competitor to AWS. Certain companies will still like to have an in house server and those would get a Mac Pro or Mac Mini but for the really large enterprise ones having Apple taking care of data security and availability would probably prove to be a much more cost effective solution provided that Apple manages to convince them that the NC centre is indeed secure, reliable, accessible, and cheap to hire.
  • Reply 50 of 101
    Like I said before, the best way to do this would be to create a wholly owned subsidiary for business sales outside the reach of jobs and Ive. This company would be able to create types of computers that the enterprise markets want, but does not fit within the typical Apple schemes.
  • Reply 51 of 101
    Quote:
    Originally Posted by Firefly7475 View Post


    ZFS would be a better solution than RAID for home users...









    Hell yeah x10. I've been going on and on about this one for ages and to be honest I can't believe they haven't announced something!



    I'm not sure about calling it Home Server though (more on that below), maybe more like iHub or iHome.



    Cloud storage is certainly the future, but we aren't quite there yet. A far better solution is to have a local device that manages content, allows the synchronization of content between devices, and then have parts of that content synced to the cloud.



    Microsoft already have similar product called Windows Home Server which, after a couple of service packs, was bloody awesome. I've played with the WHS Vail preview as well which is even better.



    Microsoft are missing out on a few key features IMO.



    Storage isn't transparent enough. One needs to be very specific about where things are being saved to ensure everything works the way it is suppose to.



    It doesn't handle the synchronization of multiple users data to the cloud. It will act as a central file store, and it can sync that file store to one SkyDrive account, but it can't sync the data from different users to different SkyDrive accounts (although this is somewhat mitigated by the fact that users can connect to the server remotely and access their files)



    The most important feature that it is missing is that it is still a PC and is treated like one. It needs to be more like an appliance. This is going to sound silly, but a PC server at home isn't sexy enough for mainstream users.



    I agree with almost everything you say -- except ZFS.



    I played around with the "pre-release" a while back! ZFS is great from a feature/capability perspective -- but while I was playing I tried to imagine how ZFS could be explained so that a non-tech user could use it, and what a GUI would look like. IMO, this would be a major undertaking.



    I envision a home server as providing: mass storage, consolidation, staging, streaming and serving files/content to multiple computers, iPads, iPods, iPhones and AppleTVs in the home or small business.. It would also provide TimeMachine style backup locally and to the cloud (syncing and staging).



    i kinda' do this now, with a Mini with 2 2-Terabyte external FW drives (soon to increased if no home server solution is offered).



    The Mini is headless and in the den ethernet-connected to my main computer for fast file exchange,



    Its main job is a media center and streams A/V content to the AppleTV and iPads (StreamToMe).



    As a file server, most of the things on the Mini are overkill or not used -- 2.26 GHz Core 2 Duo, 4GB RAM, GPU, 320 GB HDD, ODD, BT, USB, etc.



    Basically, all it does is retrieve files from the external HDDs and serve them over WiFi -- along with occasional addition of new files, TimeMachine Backup and syncing.



    So, I look at this $800 "media server" (excluding external HDDs) and this $100 AppleTV...



    Hmmm... The AppleTV gets rid of all the things I don't need/use on the Mini.



    I wonder If I got another one, hacked it and hooked the external HDDs to it via USB...



    Obviously, I am not going to do that -- I want less fiddle-work, not more.





    But, Apple could!



    How about:

    -- the guts of the AppleTV on a pluggable card, all solid-state - we'll call it a server module

    -- pluggable HDDs and/or SSDs

    -- pluggable power supplies

    -- an enclosure that accepts multiple of the above

    -- OS X Server Home (yech, sorry) -- to control it all

    ---- Cluster, load-balancing, fail over (under the covers)

    ---- Remote notifications / Admin via Mac, iPad, iPhone

    ---- Use OpenCL to exploit the GPU on the Server Modules



    What we'd have is something like:



    $150 - 2 Server Modules @ $75 each

    $400 - 2 2TB HDDs @ 200 each

    $100 - 4 Power supplies @ $25 each

    $100 - 1 Enclosure

    ====

    $750 4 TB Home Server



    Everything would be hot-swappable -- The Server Modules and Power Supplies would be expendable (recycle or Green dispose).



    This would be better, cheaper, greener than the Mini, and cost $50 less --with the HDDs thrown in free!



    .
  • Reply 52 of 101
    I haven't read all the comments but Apple and Intel have invented two Light Peaks!

    1 light peak

    2 Intel's propitiatory Ethernet/sata light bus ( name?) that will change all computers and servers and by extension supercomputers. When it enters the market I believe that servers will shrink even more than Xservers. Intel's buying of virus software company is taking over Microsoft's responsibilities and doing a "Apple" in leading PC users to the promised land of virus freedom. The first product I predict will be a super fast and super small server.

    The 40 (48?) core CPU is perfect for a server, with multi light buses to many hard drives.

    So Jobs with his inside running has seen the writing on the wall. and stopped making its server.
  • Reply 53 of 101
    He-he. The third page of the piece clearly shows Daniel's vocation is office automation and network administration, which everyone knows it is. Then how come Daniel presents pictures, which explain the evolution of iLife and iWork products in connection with server architectures? ``From your training and experience' ', Daniel, could you recall one single chance to meet that software in your professional life?



    There goes a commonplace which everyone and their dog know. Everything was crystal clear at the time of dropping cluster-grade file system support.



    Quote:
    Originally Posted by AppleInsider View Post


    Another App Store…



    Which leads to what Apple likely may be doing with Mac OS X Server (beyond simply paring it with the Mac Pro and Mac mini): taking a lesson from iOS and creating a Server App Store. By setting up a secure market for server applications, Apple could bring its successful experiment in creating a market for mobile software to the server realm.



    Businesses don't go buy servers in supermarkets; much less to any boot camps. They prefer close intimate relations with suppliers. Which in particular means maintaining a mob of office automators and help desk officers. Steve & Tim are long sniffing each single productive engineer before hiring. Investors will kill them for that gang of idlers.



    Quote:
    Originally Posted by AppleInsider View Post


    Lots of open source server software already exists, but faces difficult problems related to installation, maintenance and updating for security. By converting this software into easy to manage modules that are easy to buy and install, Apple could revolutionize server software,...

    Rather than bundling all of the company's server related projects into a general purpose Mac OS X Server package, Apple could sell its Wiki Server, Podcast Producer and Xgrid as installable modules...

    It could allow anyone to build custom packages...



    Delegating server hardware

    ...



    Apple is not software company.
  • Reply 54 of 101
    Nice idea with the server app store, although MS have already done this with windows.



    Nice chart showing how much cheaper the OSX server would be, except it's left out the fact Linux would still be cheaper.
  • Reply 55 of 101
    Quote:
    Originally Posted by Dick Applebaum View Post


    I agree with almost everything you say -- except ZFS.



    I played around with the "pre-release" a while back! ZFS is great from a feature/capability perspective -- but while I was playing I tried to imagine how ZFS could be explained so that a non-tech user could use it, and what a GUI would look like. IMO, this would be a major undertaking.



    Maybe... but I think Apple would be able to hide the underlying implementation from the user.



    Then again, the main idea I was thinking of for ZFS was swapping out drives with drives of greater capacities (WHS or Drobo style) which on further thought is actually something that would confuse mainstream users.



    i.e. It would make more sense if your iHome 2TB Module failed that it would direct you to purchase another iHome 2TB Module from the Apple Store.





    Although I like your idea of a modular server (like your own personal Blade Server!) I'm not sure non-tech types could get their heads around it. They need to be able to replace HDD modules, so they should be able to add server modules as well... I'm just not sure I like the idea of needing to buy certain components to make features work properly.



    It might be better to simply include the hardware needed to provide a list of functionality to a certain number of devices and leave it at that.



    Then again I change my mind on how they should do the hardware weekly!



    That said, the list of functionality is pretty much agreed upon by everyone I talk to. Media/iTunes storage and streaming, shared TimeMachine and data sync between all devices and the cloud.



    If Apple can nail that functionality (regardless of the hardware ) they'll have a wining product on their hands.



    Personally I'd love to see iHome manage profile synchronization between iPads as well - so if I pick up any iPad in the house it has my configuration. Same goes for the wife and kids. That's more of a v2 thing though.



    Finally I really really hope they don't call it a "server". iHub, iCenter, iHome, Apple Unity, iCentral, iPivot, iCasa, iMesh, whatever... just not "Apple Home Server" or "iServer".







    EDIT: I forgot to mention the ability to activate/sync iDevices over Wifi. The iHome should allow for the possibility of a desktop/laptop free home.
  • Reply 56 of 101
    mgkwhomgkwho Posts: 167member
    Correct me if I'm wrong, but I'm nearly positive "iWorks" is actually called "iWork."



    -=|Mgkwho
  • Reply 57 of 101
    macrulezmacrulez Posts: 2,455member
    deleted
  • Reply 58 of 101
    ilogicilogic Posts: 298member
    I have to agree with DED on this. An iOS Server App Store is not difficult to setup, especially with it right around the corner on Lion. It would disrupt and revolutionize the Enterprise level marketplace, and since Apple has everything to gain from it, you bet they're going to change everything, again. (Sheesh I just sound like a real fanboi right now)
  • Reply 59 of 101
    Quote:
    Originally Posted by ilogic View Post


    I have to agree with DED on this. An iOS Server App Store is not difficult to setup, especially with it right around the corner on Lion. It would disrupt and revolutionize the Enterprise level marketplace, and since Apple has everything to gain from it, you bet they're going to change everything, again. (Sheesh I just sound like a real fanboi right now)



    Maybe... however the homicide rate would go up.



    I personally would strangle people if they started cruising the App Store and downloading and trying out applications on a production server. I'm sure the sentiment would be widespread.



    An App Store for a personal home-based server... now that's a whole different story!
  • Reply 60 of 101
    The Cube didn't fail because of the economy, it was just an ill-conceived product that got the fate it deserved. It probably would have sold great at half the price, but it was extremely uncompetitive in terms of value, it either needed to be much cheaper or much more powerful. Basically, the price was jacked way up because it prioritized form (aesthetics) over function.



    Quote:
    Originally Posted by BenRoethig View Post


    Like I said before, the best way to do this would be to create a wholly owned subsidiary for business sales outside the reach of jobs and Ive. This company would be able to create types of computers that the enterprise markets want, but does not fit within the typical Apple schemes.



    I'm skeptical that such a company could make a profit. Seems pretty likely that the xserve line's poor sales were being subsidized by macs and ipods.
Sign In or Register to comment.