[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-users] How to set up a domU to run on multiple dom0s
I want to run a multithreaded Windows binary while distributing the threads over several computers on a local network. Since the binary in itself (mplus) has no support for distributed computing, I was hoping it would be possible to distribute the the computational threads over the network through a virtual layer. I cannot see how such a windows binary could be 'tricked' into running over two separate VMs, unless Windows provides some kind of functionality that would allow threads and sub-threads to be redirected to other computers in a network. I thought this is what cloud computing was intended for. What happened to eucaluptus, enomaly or even cloudfoundry? Robin. Ian Campbell skrev 2014-02-10 14:51: > On Mon, 2014-02-10 at 13:25 +0100, Robin Axelsson wrote: >> I'm wondering if it is possible to set up one virtual machine to use >> CPU/RAM resources located on several physical machines where some of >> these resources are shared through protocols such as MPI. >> >> Say that I have 32 CPU cores on two separate physical machines so what >> I'm essentially asking is whether it is possible to set these machines >> up so that the operating system sees 64 computing cores in the virtual >> machine / domU? > No, I'm afraid it is not. Xen carves up individual hosts and can allow > you to transfer a running VM on a very coarse time scale to another host > but does not support running one VM across multiple hosts > simultaneously. > > However I see no reason why whatever higher level tools which you would > use on two non-virtualised physical hosts to provide some illusion of a > single machine shouldn't work on two VMs hosted on different hosts (or > even the same host for that matter). > > Ian. > > . > _______________________________________________ Xen-users mailing list Xen-users@xxxxxxxxxxxxx http://lists.xen.org/xen-users
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |