Slashdot Mirror

← Back to Stories (view on slashdot.org)

AMD Demos Live Migration Across Three Opterons

Posted by timothy on from the which-walnut-under-which-cup dept.

bigwophh writes "Advanced Micro Devices has just revealed to the public the first video and images demonstrating live migration across three generations of AMD Opteron processors on VMware ESX 3.5, including the six-core AMD Opteron processor, often referred to as 'Istanbul.' For those unaware of the strains in a server environment, live migration of virtual machines across physical servers is crucial to providing flexibility for managing data centers. AMD is also taking this opportunity to highlight its continued, cooperative development efforts with Microsoft as evidenced in Windows Server 2008 R2 Hyper-V, which just also happens to be available today in beta form, that adds support for AMD-V technology with Rapid Virtualization Indexing."

25 comments

  1. Nothing new? by morgan_greywolf · · Score: 1

    VMWare ESX has been able to do live migrations for a while now. I'm not sure what makes the Opteron special in this regard.

    --
    My blog
    1. Re:Nothing new? by Enry · · Score: 1

      Xen supports live migrations too. Not sure what the big deal is.

    2. Re:Nothing new? by rachit · · Score: 4, Informative

      What is new is live migration across different CPU generations.

      I believe its using some features in the newer CPUs to turn off certain features / CPUID bits such that all CPUs look the same to the guest OS / applications running on top of it.

    3. Re:Nothing new? by Domint · · Score: 5, Informative

      The crucial part is the 'across three generations' bit - I can tell you from first hand experience that VMWare ESX has problems performing live migrations across CPUs with different steppings even within the same generation, so the fact that AMD pulled it off across 3 distinct generations without having to utilize cobbed-together solutions like EVC is a pretty big deal. Or, at least it is to me.

    4. Re:Nothing new? by morgan_greywolf · · Score: 1

      without having to utilize cobbed-together solutions like EVC

      I was just about to say 'EVC' when I got that part of your post. The thing is if you are already running EVC, then it doesn't matter much. The hard part of EVC is getting the first couple of hosts running. Once that's going, EVC is cake.

      --
      My blog
    5. Re:Nothing new? by Anonymous Coward · · Score: 2, Interesting

      No, the CPU features (the CPUID bits) are masked out with ESX, your can fine-tune this in the settings of every VM.

    6. Re:Nothing new? by Natales · · Score: 1

      It's not that ESX has issues. It's that the Guest OS is flat out not designed for "something" to change your CPU on the fly, which is the way the guest sees it... In fact we (yes, I work for VMware) used to allow that back in the ESX 2.0 days, but the guest bluescreened/panicked after a while as the microcode is loaded at boot time. The only way we can pull these tricks today is by masking some specific instructions of newer chip generations vs. the previous ones. Hard, but not rocket science. Get me a guest that adapt to those changes live and you'll have what you want...

    7. Re:Nothing new? by morgan_greywolf · · Score: 1

      Get me a guest that adapt to those changes live and you'll have what you want...

      Generally speaking, Linux distros will move between different processors as long as they are the same generation...x86-64 will move to other x86-64s, for instance.

      --
      My blog
    8. Re:Nothing new? by WaxMan0 · · Score: 1

      Pretty certian the is marketing bumpf, Is this not what the AMD-v and intels equivilent (cant remember of the op of my head) allow. Realy wish i kept my sources to refernace but from what i remember the different levels of virtuilisation include emulation where the complete pc is emulated and the OS sees this as a native hardware this is the Slow method and used by older desktop virtual pc applications. Another 2 one of which I have completely forgotten the other uses a set of cpu instructions to allow the hyper-visor to present to a modified OS an interface the OS calls this and is rerouted directly to the hardware, requires different kernals if I recall available on Linux for a while and now in Windows server 2008. This model allows for the level of migration that you would find with the fully emulated version but with close to bare metal performance, most recent opterons and AM2 athalon and phenoms have amd-v while c2d+ as well as some older P4's (some pentium D's) have the intel version. Would love to be corrected but old hat tbh. and if not only because AMD took so long to finalise the amd-v instruction sets from one core gen to the next.

  2. Yay? by eln · · Score: 3, Interesting

    They can call me when they've demonstrated seamless live migration between Intel and AMD chips, not just generations of their own hardware. Nobody wants to build a large-scale cloud if they're going to be locked to one vendor forever once they get started.

    1. Re:Yay? by CannonballHead · · Score: 3, Funny

      Yeah, what's with the wanting to show how their hardware is better than Intel's, anyways.

    2. Re:Yay? by Phantom_Linux · · Score: 1

      Didn't they do that already?

      http://www.youtube.com/watch?v=EuhU6jJjpAQ

    3. Re:Yay? by Agent+ME · · Score: 1

      I thought KVM or qemu did this?

  3. Virtual machines horribly slow? by Anonymous Coward · · Score: 0

    I've never tried VMware ESX, but anytime I've used VMWare, or dealt with customers using virtual machines in there own data center it has been horribly slow. They often say they have these 8-16 core machines with gobs of ram, and allocate more then enough to the virtual machine running our software, but without fail it often runs 25-50% slower then a basic single core server with 1gb of ram.

    I realize this can greatly depend on the load of the machine at the time, but does anyone have an URL to benchmarks with a single virtual machine running under VMWare ESX vs. the exact same machine not running VMWare ESX?

    1. Re:Virtual machines horribly slow? by Anonymous Coward · · Score: 0

      Of course, it's a little bit slower - though in my experience not the numbers you've been getting. But get a cluster of VMWare ESX servers and the extra redundancy and ease of provisioning new VMs and reconfiguring existing VMs really does make up for the slowness.

    2. Re:Virtual machines horribly slow? by Anonymous Coward · · Score: 0

      Unfortunately VMware really is very very slow, even in comparison with other competing vendors software virtualization technologies. I've found it to be anything upto half the performance of alternatives. AG

    3. Re:Virtual machines horribly slow? by Anonymous Coward · · Score: 0

      Now you're just sounding like a troll.

  4. What about networks? by Anonymous Coward · · Score: 0

    This is pretty impressive. Instruction set differences and chipset differences have to all be abstracted away. No small feat.

    I was curious how they migrate active network connections though. Does the old host act as a proxy/router? Can anyone shed some light?

    1. Re:What about networks? by Chris+Burke · · Score: 3, Interesting

      I was curious how they migrate active network connections though. Does the old host act as a proxy/router? Can anyone shed some light?

      Its been around 9 years since I did the project in college, but it is possible to transfer active network connection state from one computer to another. It was a "connection-aware seamless backup server", where our hacked linux kernels would exchange state about an active TCP/IP connection at regular intervals, and when the primary dropped (in our demo we yanked the ethernet cable out of the hub in the middle of streaming an mp3), the other would take over, pretending to be the same IP and picking up where the other left off. The best part was that TCP/IP already deals with redundant, missing, or out of order packets so anything sent or received since the last update to the backup would be handled automagically.

      That was just as stupid semester project on a 3-computer ethernet LAN, but I imagine the big boys have figured out how to make it work. Besides, they're literally transfering the memory image of the guest OS over to the other machine so all the state update is already done. The hard part is probably making the IP migrate along with, but I'm sure they've figured that out too.

      --

      The enemies of Democracy are
    2. Re:What about networks? by Domint · · Score: 1

      Nothing that complex. It simply passes the TCP state from one nic to the other. I don't know the gory details of how that all happens behind the scenes, but it's pretty impressive to watch it happen - in our lab at work we tested this functionality on multiple guest OSes simultaneously and didn't lose a single packet, even when pulling the power chords out of a number of components (including one of the ESX servers).

    3. Re:What about networks? by cowbutt · · Score: 3, Interesting

      That was just as stupid semester project on a 3-computer ethernet LAN, but I imagine the big boys have figured out how to make it work

      Checkpoint's Firewall-1 product has been able to transfer the current firewall state to a backup firewall for quite some time (at least 1998 or so), but it originally required the use of RIP to implement failover which introduced delays. In about 1999, Stonesoft produced an add-on product, Stonebeat, which added ARP spoofing, so the failover was virtually instantaneous. These days, the failover is done using VSRP.

    4. Re:What about networks? by chrome · · Score: 1

      as far as the nics are concerned, its all just ethernet packets. There is no state to handle. The state is maintained by the guest OS, and is transferred when the OS memory is transferred.

    5. Re:What about networks? by omfglearntoplay · · Score: 1

      Supposedly Double-Take software is good at "mirroring" one box with another... and then when box1 dies box2 very quickly takes over, stealing the IP, etc. I have the software, but I haven't had the balls or the weekend time to test it for real yet.

    6. Re:What about networks? by Anonymous Coward · · Score: 0

      "The hard part is probably making the IP migrate along with, but I'm sure they've figured that out too."

      It's called VRRP, it's been around for a very long time. I use it to move around Xen guests and for router failover. Works great.