Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Software Businesses Novell Red Hat Software Linux

Red Hat, Novell To Package Xen 233

Posted by CowboyNeal from the brand-new-player dept.
robyannetta writes "Watch out VMware and Microsoft. Here comes Xen, an open-source virtualization for the Linux environment being pushed by Red Hat and Novell. Xen has also joined forces with leading Linux distributors, chip vendors and platform vendors to create a consortium that will more broadly enable open-source virtualization development and deployment." We've covered Xen before, but it's cool to see the momentum behind it growing, as more choice is a Good Thing.
This discussion has been archived. No new comments can be posted.

Red Hat, Novell To Package Xen More

Red Hat, Novell To Package Xen

Comments Filter:

  • Giving me a headcrab-ache (Score:5, Funny)

    by SIGALRM ( 784769 ) * on Thursday December 02, 2004 @07:46PM (#10981156) Journal
    Xen has also joined forces with leading Linux distributors
    Am I missing something or does this mean Vortigaunts might enter Earth through portals ... wearing Fedoras?

    Oh my, it all makes sense now.

  • USE THiS XEN LINK (Score:5, Informative)

    by Anonymous Coward on Thursday December 02, 2004 @07:49PM (#10981190)
    http://www.cl.cam.ac.uk/Research/SRG/netos/xen/ [cam.ac.uk]

  • Uhm... (Score:5, Informative)

    by Richard_at_work ( 517087 ) on Thursday December 02, 2004 @07:52PM (#10981219)
    'Watch out VMware and Microsoft'? If im correct, Vmware and VPC doesnt require the host operating system to be actually ported to the virtual system, whereas Xen does. This might be fine for specific usage, but its next to useless for what I use vmware for - trying out new and interesting operating systems, configurations or such. The markets may overlap near the top end, but I see no reason why VMware/VPC need watch out, as the main market for these VMs is running Windows, and while there has been a developers port of WinXP to Xen, I severely doubt you will see that in the wild.

    • Re:Uhm... (Score:5, Informative)

      by interiot ( 50685 ) on Thursday December 02, 2004 @08:00PM (#10981297) Homepage
      From the article...
      • Xen does not support Windows today because it uses a technique called para-virtualization to achieve high performance that involves modifying the operating system kernel, Pratt said. However, the debut of virtualization features in next-generation CPUs from Intel and AMD will make it easier to support unmodified operating systems, Pratt said.

      • "At that time we will reconsider Windows support," he said.

      And more here [ogi.edu]:
      • Full virtualization requires no changes to the guest OS. But it is not well suited for x86 architecture as x86 is not virtualizable. x86 has machine instructions that are termed sensitive. Sensitive instructions will fail without trap when executed in the guest OS. This requires dynamic rewrite of guest OS kernel during run time and shadow system structures which results in a performance hit. Para-virtualization solves the above issues but it requires changes to the guest OS kernel although are required for the Application Binary Interface (ABI) and hence applications can run as they are on the guest OS.

      • Re:Uhm... (Score:3, Informative)

        by interiot ( 50685 )
        And more here [nwfusion.com]
        • Virtualization technology has been used on mainframes and high-end servers for years, but IT departments are starting to use the technology on low-end servers as well

        • "In order to virtualize technologies within a processor, a little bit of hardware goes a long way," Brookwood said. Users still need virtualization software, but that software will run much faster with hardware support, he said.

          Neither Intel nor AMD has built such technology into their processors for low-end servers, but bot

        • Re:Uhm... (Score:4, Interesting)

          by interiot ( 50685 ) on Thursday December 02, 2004 @08:16PM (#10981461) Homepage
          Are there any good benchmarks out there comparing Xen to VMWare? this PDF [ira.uka.de] contains benchmarks for very specific operations, not entire programs, but indicates that Xen is much faster for those operations. Though it also notes that VMWare's license prevents people from publishing benchmarks. So... This might mean that VMWare itself realizes that they have severe performance problems in places?

          • Re:Uhm... (Score:3, Interesting)

            by drinkypoo ( 153816 )
            I have the same question, but in relation to User-Mode Linux. The assertion is made in the article that UML is slower than Xen. I don't see why it would be, because they are essentially the very same thing, and I doubt the UML developers are a bunch of potzers or anything. I have not used UML much but when I have it's been plenty speedy and very enjoyable. I'm preparing to use UML to break up my server tasks to improve security, and give my remote users a separate system they can't break too easily. I doubt

            • Re:Uhm... (Score:3, Informative)

              by interiot ( 50685 )
              It's slow because of its very name -- "user-mode". Generally anything that has to do kernel-like stuff, but does it in user-mode, will be a fair bit slower than it could be if it were in the kernel. As I understand it, this is because every low-level operation (eg. sockets, disk I/O, etc) requires memory to have one extra copy made for each transaction.

              Presumably things like microkernels [wikipedia.org] get around this by sharing memory directly with the kernel? Though that article says that microkernels still histori

              • Presumably things like microkernels get around this by sharing memory directly with the kernel? Though that article says that microkernels still historically are slower than monolithic kernels, so the user-mode thing might clearly be an unresolvable problem here.

                IANAT (I am not Andrew Tanenbaum) ;-)
                Basically a microkernel that shared memory between processes would not be a microkernel. The page you linked to had another category:

                Hybrid kernels (modified microkernels) [wikipedia.org]
                Hybrid kernels are essentially

            • Re:Uhm... (Score:3, Informative)

              by l810c ( 551591 ) *
              I've just recently moved from UML to Virtuozzo VPS and the speed increase was dramatic given the same hardware specs.

              Here's [tektonic.net] a post from my current VPS provider that does a great job of explaning all 3.

          • how the hell can a company prohibit some reviewer from publishing benchmars?? THERES NO WAY THAT'D STAND UP IN COURT.. that's like a car manufacturer prohibiting a reviewer from revealing the cars braking performance or something...

      • Re:Uhm... (Score:3, Informative)

        by Richard_at_work ( 517087 )
        Interesting, because the Xen homepage [cam.ac.uk] has this [cam.ac.uk] to offer:

        1.3 Which OSes run on Xen?

        To achieve such high performance, Xen requires that OSes are ported to run on it. So far we have stable ports of Linux 2.4, Linux 2.6, and NetBSD. Ports of FreeBSD and Plan 9 are nearing completion.

        1.4 Does Xen support Microsoft Windows?

        Unfortunately there are no plans to support any versions of Windows in the near future. Furthermore, a port of Windows would be encumbered by licensing issues. Longer term, virtual

    • Re:Uhm... (Score:4, Informative)

      by mccalli ( 323026 ) on Thursday December 02, 2004 @08:04PM (#10981354) Homepage
      Watch out VMware and Microsoft'? If im correct, Vmware and VPC doesnt require the host operating system to be actually ported to the virtual system, whereas Xen does.

      You're right. However, for once people are using their terms correctly whereas they normally get mixed together.

      Virtual PC, despite the name, is not virtualisation software. It's an emulator - has the whole chip and other bits of hardware under there to run, even if it's natively running on an x86 anyway. That's why it's useful to me over VMWare, as I swap the same virtual machine between Mac and Windows platforms.

      VMWare is virtualisation software. It doesn't emulate as such, instead it provides hooks to access the native platform as if it were a separate environment. That normally makes it quicker than an emulator, and I believe this is normally borne out in various speed comparisons with Virtual PC.

      I've not encountered Xen, but from how things sound it really is a proper virtualisation package and not any form of emulator. It sounds like it is providing kernel hooks to access its current Linux environment as if there were multiple environments. So it definitely is virtualisation. Think of IBM's zSeries virtualisation - that needs special coding too, from what I recall.

      "I see no reason why VMware/VPC need watch out, as the main market for these VMs is running Windows..." may be how most people think of it, but Virtual PC is not a VM - it's an emulated environment.

      Cheers,
      Ian

      • Re:Uhm... (Score:3, Informative)

        by homer_ca ( 144738 )
        Umm... VirtualPC is both. Virtual PC for the Mac is an emulator that translates x86 code to PowerPC. Virtual PC for Windows is a virtual machine that executes x86 code natively.

      • Re:Uhm... (Score:5, Insightful)

        by lakeland ( 218447 ) <lakeland@acm.org> on Thursday December 02, 2004 @08:34PM (#10981587) Homepage
        Xen won't run an OS unless that OS has been especially ported to Xen. I.e. it is ideal for running linux, BSD, and not much else. It is also x86 only (I can't use it as a MOL replacement).

        It has one huge benefit over VMWare, it is extremely fast. The virtual machine has so close to the performance of the host that it would be reasonable to do such things as: implement a 100% reliable server on your computer and then implement an up-to-date desktop machine inside it. Implement virtual hosting on cheap x86 hardware. Run two distros simultaneously, etc.

        Personally I think running a reliable server on the same hardware as your unreliable desktop would be nice. Have the one machine always work correctly for handling mail, printing, web serving, etc. But still up to date.
        • Aren't the virtualization projects like Xen, User Mode Linux, etc. essentially reinventing the microkernel? Their goals seem awfully close to the idea of abstracting away the hardware so you can safely run multiple OS "personalities" at the same time. I wonder what could be accomplished if the virtualization guys teamed up with the people working on a modern microkernel like L4 [l4hq.org]. Anyone have benchmarks comparing L4-Linux [tu-dresden.de] with Linux under Xen? Heck, maybe one day people will realize the advantages of a mul

        • "It has one huge benefit over VMWare, it is extremely fast. The virtual machine has so close to the performance of the host that it would be reasonable to do such things as: implement a 100% reliable server on your computer and then implement an up-to-date desktop machine inside it. Implement virtual hosting on cheap x86 hardware. Run two distros simultaneously, etc."

          Obviously you haven't run VMWare lately.

          Anything running in the version 4 series of VMWare (now up to 4.5.2) runs as fast (and faster in

        • Cambridge University are source licensees for Windows and they have a modded HAL that can handle the interface with the VM Monitor. They are not, under the terms of their license, permitted to distribute the binary.

          However, it is possible that MS themselves may want Xen support over time. The dinosaurs have shown us that it is advantageous for an O/S to share a platform and it can give significantly more flexibility.

      • It is worth mentioning (and you did not, so I am) that vmware emulates everything but the CPU and USB devices. It's unfortunate that they don't have a scheme for a driver to allow you to directly use installed hardware by preventing windows from using it and giving you access to all its resources. You would then load a driver to access it in your virtual machine. Shit, even dosemu let you have direct access to hardware devices in your system that were unclaimed by Linux.

        In that respect vmware both emula

    • "Porting" (Score:5, Interesting)

      by vlad_petric ( 94134 ) on Thursday December 02, 2004 @08:13PM (#10981437) Homepage
      That's a little bit overstated. For Linux, the changes are a couple thousand lines of code. That's a number I took from their research papers.

      But what do you have to change ? First of all, the system has to be made aware that it's not the "top top". Its physical memory is no longer contiguous (you ask Xen for memory pages, and it gives them to you in arbitrary places), it also has to be aware of absolute time that's no longer tick++. Second, you need drivers for the abstract network card and disk. Those are generally easy to write, because you just delegate the real work to Xen. VMWare is already doing something similar with its vmxnet driver for Windoze.

      I'd really expect these kind of changes to the OS to be incorporated in the main linux tree, as they mature.

      What do you gain from all this ? Well, SPEED. I mean, SPEED. Take a look at their research papers (wrong suggestion for the "I won't RTFA" crowd, but still ...). Their slowdowns/throughput losses (they run Postgres and Apache on a couple of virtual nodes, as opposed to a single, consolidated machine), are negligible (less than 10%). On some configurations they even got performance improvements! At the same time, VMWare and UML do considerably worse.

      In general, it's very easy to "virtualize" stuff that's running mostly in user space. As soon as you have considerable OS+I/O overhead, your performance drops significantly. The para-virtualization approach (employed by Xen), pretty much gets you the best of both worlds.

    • is, as the article states, business server use. Put the VM images on a SAN, run the virtualization software on the real hosts.

      Then when you need some CPU power, add a real host, suspend a VM, and resume it from the new real host.

      Last time I checked, all servers don't run Windows.

      VMWare is an ugly hack, that will ultimately perform worse than things like UML and Xen because of the unnecessary requirement to emulate a CPU by Windows.
    • If im correct, Vmware and VPC doesnt require the host operating system to be actually ported to the virtual system, whereas Xen does. This might be fine for specific usage, but its next to useless for what I use vmware for - trying out new and interesting operating systems, configurations or such

      I believe the majority of people who buy VMWare, though, are using it to isolate servers. E.g., if you have two departments or customers who both need a web server, and don't each need the full resources of a ded

  • Xen [sf.net]
  • From TFA:

    Xen does not support Windows today because it uses a technique called para-virtualization to achieve high performance that involves modifying the operating system kernel, Pratt said. However, the debut of virtualization features in next-generation CPUs from Intel and AMD will make it easier to support unmodified operating systems, Pratt said.

    As usual slashdot is overhyping or just getting shit plain wrong in article summaries. This is yet another usermode linux clone it seems. This is probabl
    • Well, there's QEmu [bellard.free.fr], and Bochs [sourceforge.net], which can run Windows in Linux and so on. However, they are full CPU emulators, and are as such much much slower than VMWare. If all you're using VMWare for is like Outlook in Windows (and some organizations I know do exactly that), then these solutions might suffice, but for anything else VMWare wins hands-down.
    • Xen is *very* different than plex86. Xen is a virtual machine monitor - it directly executes most instructions, and achieves performance that is within a few percent of non-virtualized operating systems.

      Plex86 is an emulator - it interprets most instructions, and it is dog-slow.

      It's true that Xen requires the guest OS to be ported to the Xen virtual architecture, but this has been done for linux.
    • In server consolidation environments, Xen is very much a viable alternative to VMware ESX -- it's faster and cheaper.
    • Really, I think much of the "virtual OS" market is going to remain rather stagnant, because commercial OS makers have little real interest in it.

      Think about it... If you're Microsoft, for example, and you think it's worthwhile for Windows XP to run well on say, a Macintosh system, you wouldn't expend loads of effort getting it running quickly inside an emulator like Virtual PC. You'd just release a native version of "Windows XP for Macintosh"! (They could even make it "play nice" with OS X so you'd have
    • I have played with Xen and looked at the source. The issue is the impossibility of hardware traps when privileged instructions are run outside ring 0. Xen replaces these instruction with assembly calls into a Virtual machine monitor, a modded version of Linux which acts as the host. As long as guest O/S systems can be modded to invoke the VMM rather than execute instructions directly, it works fine. This requires source code mods.

      The issue with VMware is that it is at best a hack. It tries to locate and m

  • I already use VMWare, and while I love the concept, I have had several problems, especially in using it to test newer versions of OS's. (Fedora Core 3, for example, could not load the kernel properly, while it wouldn't even recognize the disc for Fedora Core 2. Yes, I checked the checksum, yes, it matched.)

    It would be great if someone could come up with a better (and free) alternative to it, hopefully some of these bugs can be worked out. I would certainly like to see all the "good" features kept, such as

    • I have installed and run Fedora Core 3 (and before that FC2) in VMware Workstation 4.5.2-8848 under WinXP (urgh) at work. I have had no problems at all. When I compiled 2.6.9 in FC2, the performance increase inside VMware was very noticeable.

      If you want my VM files (minus the disk) and xorg.conf, contact me via my site.

      For those trying to run VMware Workstation on Fedora Core 3, this is in the release notes:
      VMware WS 4.5.2 is known to work on Fedora Core 3 after the following workarounds are used:
      *

  • Watch out, here it comes. An "open-source virtualization" something or other. WOW! Is it also robust and oriented toward our needs? Neat! Moles and trolls, moles and trolls.
  • Xen is a Good Thing (TM).

  • Xen is good stuff (Score:4, Informative)

    by gbnewby ( 74175 ) * on Thursday December 02, 2004 @08:04PM (#10981352) Homepage
    I was an early user of 1.0, and have followed developments of Xen since. It's very nicely done, open source, and builds on existing kernels and distributions (it's not really a standalone application, but integrates with the Linux kernel and adds some userland tools).

    Xen lets you configure one physical system with multiple virtual systems. Hardware access (disk, net, video) is transparent via software.

    This is kind of the conceptual opposite tools like Condor and Globus: rather than bundling lots of physical systems together as one (aka, grid computing), it is meant to take one system and subdivide. This makes for easier development (including testing for grid services, Web services, different distros, etc.), and of course is good for virtualization (like in Web hosting services).

    Congrats, team!

  • dont see it in FC 2 yet (Score:2, Informative)

    by argoff ( 142580 )
    [root@ root]# yum install xen #yeah like this will really work
    Gathering header information file(s) from server(s)
    Server: Fedora Core 2 - i386 - Base
    Server: Fedora Core 2 - i386 - Released Updates
    Finding updated packages
    Downloading needed headers
    Cannot find a package matching xen
    No actions to take
    [root@ root]# find /lib/modules/2.6.9-1.6_FC2/|grep -i xen
    [root@ root]# find /var/cache/yum/|grep -i xen
    [root@ root]#

    • That's because the latest features appear in the Fedora Core development tree, not in already-released versions.

      • That's because the latest features appear in the Fedora Core development tree, not in already-released versions

        Does that mean that it eventually will be released into fc2?... also I'm not quite sure how it will be released, as another kenrnel, package, or module? Right now I'm using a manual build on top of FC2, but it does not work too nicely with iptables, for example.

  • What about Qemu ? (Score:3, Interesting)

    by Anne Honime ( 828246 ) on Thursday December 02, 2004 @08:07PM (#10981372)
    I'm a happy Qemu user on fedora 2 ; although not as flexible as plex86 wrt host architecture, nor as polished as VMware, Qemu has windows support, and just does the job fine, after minor tweakings.

    I've yet to try Xen, but as of now, I just need basic windows compatibility to launch closed softwares (most of them being databases of law articles on CD-ROM to copy / paste extracts in linux Openoffice for research purpose), and Qemu does just that.

    I only wish it could play GTA3-VC !

  • Novell's Zenworks, which rendered 1000 NT4 PC's/night useless on a regular basis while I was in college.

  • What is Xen? Here you go... (Score:5, Informative)

    by Erich ( 151 ) on Thursday December 02, 2004 @08:23PM (#10981522) Homepage Journal
    Thought everyone might like an explanation of what Xen does, in comparison to VMware.

    VMware works with a host operating system to provide a complete x86 virtual environment for a guest operating system.

    Xen is an operating system in its own right. It's a "virtual machine monitor" or "hypervisor". It can spawn multiple guest operating virtual machines.

    x86 is not a very good architecture for virtualization. To have a virtualizeable architecture, anything a user-level program can do should behave the same way it would in supervisor mode, or it should trap so the virtual machine monitor can emulate it. x86 has instructions that don't quite follow this guideline -- for instance, you can see what protection ring you are currently in. In supervisor mode, you would get something like ring 0. In user mode, you get ring 3. So an operating system trying to see what ring it was in would get ring 3, but you are trying to fool it into thinking it is in ring 0.

    Anyway, Xen modifies the guest architecture. It disallows these "sensitive" instructions and creates some virtual devices that are easier to emulate (like a simple software-programmed TLB). This allows the performance to be very very good, faster than VMware, but it requires you to fiddle with your operating system a bit. Which, of course, is easy to do with Linux.

    • The terms 'OS' and 'monitor' bug me...

      Xen doesn't seem like an OS by any traditional sense of the word. If it was, it could be run with just some firmware installed, rather than having ports to OSs like Linux and BSD.

      Them describing it as a 'monitor' doesnt' make much sense, as it seems to have a virtualization engine itself (the term monitor makes it sound like you need other software for that).

      'Hypervisor' is a wank.

      How about:

      "Xen is a virtual machine system, allowing you to run various guest operati

    • History lesson... (Score:4, Insightful)

      by Bozdune ( 68800 ) on Thursday December 02, 2004 @09:54PM (#10982175)
      Back in the 70's, CP/CMS on the 360 architecture was open-source. CP originally was a pure emulator: privileged machine instructions caused interrupts and were then emulated by CP. So we used to debug entire CP/CMS operating systems on top of CP, which was pretty cool. IBM eventually released a commercial version of CP/CMS called VM370.

      Meanwhile a bunch of independent companies, mostly time-sharing outfits, figured out that emulating privileged instructions was, well, dumb and slow. Instead, if you modified CMS to use traps instead of executing emulated privileged instructions, it could run many times faster.

      Which is why commercial timesharing outfits like National CSS, etc., were routinely able to support 250+ users on 370/168's, roughly three times the user load that IBM could support with VM370. That, and the fact that National CSS bought up every single drum drive they could find as paging devices. Ridiculously fast for the time -- nearly zero seek time, and delightfully high RPM's -- but when the bearings froze, those suckers would often burst right out of the glass case and blow holes in concrete walls.

      Anyway Xen is not a new idea. It's a very old (and good) idea.

  • I just hope it doesn't suck like the OTHER Xen.
  • I think something that folks are missing here is that Xen is not some whizzbang way to run your favorite Windows programs or to try out new versions (or completely different) of an operating system.

    Xen is more similar to VM; it's already great for server farms and when things like OpenSSI become available/usable it'll mean the realization of network-wide clustering that's Tannenbaum's wet dream.

    VMware/Qemu/Plex86 might be [aguably] good for those who only pretend to use their computer, but they're absolut
  • Intel is actively working on adding hardware virtualization (Vanderpool) to its lineup of x86 processors. This will make products, like VMWare, obsolete.

  • Hmm (Score:2)

    by T-Ranger ( 10520 )
    I wonder if Novell will create a management system, allowing point-and-click and/or programmable provisioning capabilities. They could call it ZENWorks for Xen! The manual could be called "Zen and the Art of Xen with ZENWorks"

  • Virtualization (Score:4, Informative)

    by gtrubetskoy ( 734033 ) * on Thursday December 02, 2004 @08:58PM (#10981754)

    There are 4 ways (I think) to provide what is loosely referred to as "virtualization":

    1. Hardware emulation. QEMU, VMWare, Bochs all fall in that category. QEMU is open source and is actually pretty cool - a great way to test kernels during development or testing that new ISO you're trying to put together. This method is the slowest of all since all hardware is simulated in software.

    2. User Mode Linux. In this scenario the kernel is run as a user process. This method has the second most overhead. Security-wise, it is only as secure as the host system, so if there is a known userland exploit, it is vulnerable.

    3. Xen. To the best of my understanding, Xen is a kernel which runs other kernels. So this architecturally similar to UML, but (if you believe them) is much better optimized. And if Xen is as exploit-free as is claimed, it should also be pretty pretty secure, though I believe only time will tell.

    3. Separation. This is Linux VServer, which is a fantastic project that doesn't have the publicity engine and funding of a big university behind it. This isn't really virtualization as much as it is separation. This approach is also shared by SwSoft's Virtuozzo, FreeBSD jails and Solaris containers. Since there is only one kernel in this scenario, this method is not OS-independent, i.e. VServer only runs Linux, Jails are only for FreeBSD, etc. Performance-wise, this approach should far outrun any other method as it carries practically no overhead and takes advantage of all the existing UN*X optimization. It is also very secure, possibly most secure of all (short of hardware emulation like QEMU) since it directly addresses all known virtualization exploits such as chroot escapes. But, perhaps I'm biased...

    • Re:Virtualization (Score:5, Informative)

      by IntergalacticWalrus ( 720648 ) on Thursday December 02, 2004 @09:12PM (#10981869)
      VMWare is NOT an emulator, it is a virtual machine. x86 instructions are run natively with some magic to fool the kernel into thinking it is having control of the CPU. Think Xen without the necessary kernel hacking.

      The Windows version of Microsoft Virtual PC is a virtual machine, too, while the Mac version is, quite naturally, not a virtual machine but an actual hardware emulator since it runs a different target machine than the host. (Yes, they did give two completely different products the same name.)

      Another virtual machine, but running on PPC instead, is Mac-On-Linux.
    • [Single-kernel separation] is also very secure, possibly most secure of all (short of hardware emulation like QEMU) since it directly addresses all known virtualization exploits such as chroot escapes.

      But the separation architecture does not address unknown exploits such as kernel bugs. And any kernel as big as Linux+VServer does have bugs. In Xen, the guest kernels are themselves confined, so kernel bugs are not security-critical. In Xen, you only have to worry about bugs in the hypervisor itself, and th

      • In Xen, you only have to worry about bugs in the hypervisor itself, and the hypervisor is much smaller than Linux.

        Yes, but "smaller" may be means "less work to secure" (source code clarity discipline is a big variable here), but does not automatically mean "more secure", which seems to be the claim. And in "less work" we're probably talking years. The linux kernel has been around much longer and is reviewed so much more widely. I'd be very careful claiming that Xen is somehow magically by design impene

  • I see a distribution paradigm (Score:5, Interesting)

    by argoff ( 142580 ) on Thursday December 02, 2004 @09:02PM (#10981774)
    IMHO, this cound change the entire distribution paradigm - for example - if you wanted an email server - you just download the virtual image off the network of a linux system that already has it preinstalled and mostly set up, of if you want a web server/ldap server/dns server - same thing. It is a very nice way to have the best of a full featured linux system while at the same time the xen application os has the minimal stuff nesissary to run what you want.
    • This sounds almost like my experience with an mac os 9 server(here me out).

      OS 9 was actually quite nice, for software maintenance and configuration. Very consistant and either there were no installers, or the installers were just smart copy/paste scripts.

      I had an OS 9 box that I wanted to run as a web server. I found one, download the file, doubl-clicked it and pretty much without any more effort a web server was up.

      Granted, the whole reliability/multi-tasking issues for OS 9 make it a less than ideal sy
      • I know, different way of getting to your solution. It isn't as secure or 'cool'. But the self-contained binary for the software I wanted was really impressive.

        The Big Binary File (BBF) sounds nice from the standpoint of getting started. It's all there, point, click, away you go. What could be better than that?

        BBF's are nice for little stuff that doesn't matter much, but quickly become very sucktastic once you start to scale up.

        For example, what happens when a security hole is found in a library used in
  • I read some papers on Xen. it is certainly *not* the virtual machine to start with. It won't take unmodified version of Linux and run it. Extensive changes are required. The major benefit is high efficience compared with the 'classic VM' approach.

    Given that, there is no reason why tweaks required by Xen can't be made the standard feature of the Linux Kernel. In fact, the major problem with any VM approach - either it is not efficient, or it requires some kernel tweaks. Essentially, the 'host os' and the 'g
  • Why all this hype about Xen? Is there really a significant market for running multiple open source operating systems simultaneously on the same machine? Last time I checked, the greater majority of uses of a virtual machine is either running Windows on another OS, or running another OS under Windows... But maybe I'm missing something cool (and profitable) about this?

    • Re:Target audience? (Score:4, Informative)

      by iggymanz ( 596061 ) on Thursday December 02, 2004 @09:56PM (#10982193)
      With ESX VMware is making money on consolidating many underutilized servers onto one box (with redundant failover box if needed) without the fear that a bad app misbehaving & killing the OS on one virtual machine will lock up other virtual machines' OS. It also provides customizable virtual network(s) between the VMs. A true open source equivalent would be very cool (one that can run ANY i386 OS in VM) You can migrate virtual machines from one physical box to another, sync, and cut over without interruption if they're both on same subnet. Kewl, hope a true open source equivalent exists someday.
    • Nope. The greatest (from a business perspective) number of users of virtualization technology is in the enterprise market. That's the biggest reason people by systems like the z-Series--consolidation. The idea is that you consolidate a whole bunch of commody machines to one large system that's virtualized. You get the same network topology at the end of the day but a higher degree of reliability and a lower administration cost (hence lower TCO).

      It's an amazingly powerful bussiness case.
      • If a hacker gets root in a hole in Apache, only the VM instance gets compromised. Its alot like the BSD jails but more popular.

        What I like about it is when I take my assembly course next year in college I can run it in a VM state. Assembly is a great way to freeze up and fuck up your computer. With Vmware or Xen I just type the code in Vim and cut n paste it in the vm session running Linux and execute the code. If it freeze its a no biggie and I just restart the session. No long reboots and lost saved work
  • because it is not general purpose such that I can download and experiment on any OS I want, but I am restricted to "specially modified" version of them.

    I really don't care about Windows, I want to run Darwin x86 or FreeBSD, and not just "specially modified" versions, but ANY version I so choose.

    Xen cannot handle this, and that, in my view and experience, makes it relatively worthless when compared to something like VMWare. Attempting to sell the lack of a hardware virtualization layer as an "advantage" i
  • I suspect that you could accually port NT/W2k/XP to this pretty easy without M$'s help. That is if you have a copy or can get a copy of the NT HAL development kit from M$. The kinds of operations that XEN is requiring OS changes for are already in NT as part of the kernel arch, Page insertions, TLB flushes, etc all go through common linked in routines. Unlike in linux where the routines are compiled into the kernel, NT makes calls to HAL.dll which gets linked by ntldr. Accually it probably wouldn't be to ha

    • "if" (Score:2)

      by bani ( 467531 )
      That is if you have a copy or can get a copy of the NT HAL development kit from M$.

      that's one pretty damned big "if".
    • Cambridge has a source license for XP and is very closely associated with a Microsoft research lab nearby. They did the port but unless you are a) an academic and b) also a source licensee, you won't get the source patches and don't have a hope in hell of getting the binary.
  • Does anyone around has used Xen on a real production environment like, say, a VPS provider? What About Linux Vserver (http://www.linux-vserver.org/ [linux-vserver.org])?

    What are the real differences, besides the technical paravirtualization of Xen and the fact that the guest OS must be 'ported' to it in order to run it, and that Linux Vserver is for running Linux only? I mean in terms of performance, feature-richness, security, stability and scalability of both the host and guest OS? What about work under non-x86 architecures

  • Xen is the real deal. (Score:5, Interesting)

    by kma ( 2898 ) on Friday December 03, 2004 @12:41AM (#10983290) Homepage Journal
    Disclaimer: I work for VMware, but they don't pay me to post on slashdot.

    There are a lot of replies of the form, "Wait a minute, Xen requires that you hack up your guests! What a crock! Typical slashdot hype!" It's true that Xen is more limited than VMware's products, in that you do need to modify guests. However, this doesn't mean that Xen is a joke. (Plex86, for instance, really is a joke, because Kevin Lawton seems to pursue it only in fulfillment of an elaborate VMware-centered revenge fantasy.)

    The Xen folks, on the other hand, are smart and mostly serious people. Xen, along with appropriately modified guests, solves some of the problems that our products solve, and for those areas where it fits the bill, it does so in a way that should have lasting performance advantages over full x86 virtualization. What Xen is not, in my opinion, is a virtual machine monitor, for any reasonable definition of VMM. Xen is a microkernel. They don't call it that, because it's hard to get papers about microkernels published these days, but if you think about it, the process of porting an OS to run as a guest under Xen isn't cocnceptually distinct from porting it to run as a personality under Mach or Chorus or whatever. The L4 [l4ka.org] people didn't even bother renaming their microkernel before repurposing it as a paravirtualization platform.

    I think the microkernel analogy helps clarify ones thinking about the promises and limitations of so-called "paravirtualization." Hypervisors are microkernels. In the mid-90's, there was a hope that the whole world would be able to settle on the Mach microkernel. It never happened. Anybody hoping to become the only 'para-hypervisor' will face the same political and commercial challenges.

    So to recap: Xen is not a replacement for VMware's products. Xen will probably not take over the world to the degree that its creators would like. Xen is not, however, a joke. The Xen researchers are mostly conscientious, smart people who, fairly enough, would like to see their work have some commercial impact. I really wish they'd stop beating their chests over benchmarks that show them beating a three year old version of our desktop product, though.

    • Re:Xen is the real deal. (Score:5, Insightful)

      by Ivan the Terrible ( 115742 ) <vladimirNO@SPAMacm.org> on Friday December 03, 2004 @02:47AM (#10983822) Homepage
      I really wish they'd stop beating their chests over benchmarks that show them beating a three year old version of our desktop product, though.

      Does VMware's license forbid its use for comparison purposes? If so, it's up to you (VMware) to change it. The ball is in your court.

      If not, why don't you give them a copy to benchmark with? It's not like you'll lose a sale, so the out-of-pocket cost is effectively zero. In fact, it's probably a net gain because of the less-than-favorable publicity they generate. Again, the ball is in your court.

      Or, if can't or won't give the s/w away, why don't you publish some benchmarks? Yes, again, the ball is in your court.

      Let us (the Greater Slashdot Community ) know what you plan to do.

      • Re:Xen is the real deal. (Score:3, Informative)

        by kma ( 2898 )
        Does VMware's license forbid its use for comparison purposes?

        The blanket license does, though we've made exceptions when researchers ask nicely. See, for example, Marko Zec's OASIS workshop paper from ASPLOS XI [tel.fer.hr], which includes benchmark comparisons against a reasonably recent version of VMware Workstation (that show Workstation in a pretty unfavorable light, I might add). I can only speculate as to why the Xen folks don't get treated as well as Marko did; I don't even know for a fact whether they've asked

    • Re:Xen is the real deal. (Score:5, Informative)

      by Doodhwala ( 13342 ) on Friday December 03, 2004 @03:15AM (#10983906) Homepage

      The Xen researchers are mostly conscientious, smart people who, fairly enough, would like to see their work have some commercial impact. I really wish they'd stop beating their chests over benchmarks that show them beating a three year old version of our desktop product, though.

      All right... so while I accept most of what you have said earlier as quite informative, I do take strong objection to the above statement. You do realize that the research community is forced to benchmark against Workstation 3.x because your EULA in later versions prevent any of us from publishing benchmarking numbers (Look at the Restrictions section in the EULA [vmware.com] for Workstation 4.x).

      While I understand that there might be commercial reasons behind it, it seems that VMware wants to play in the research field (publish papers at all the top systems conferences) but not allow anyone to try and reproduce what their research claims.

      Disclaimer: I have worked on virtualization projects including Xen.

Slashdot Top Deals

"The vast majority of successful major crimes against property are perpetrated by individuals abusing positions of trust." -- Lawrence Dalzell

Close