Ulli Horlacher <frams...@rus.uni-stuttgart.de> wrote: > A small network application benchmark between LXC and VMware ESX: > > ESX: > > framstag@diaspora:~: fexsend -i unifex /tmp/2GB.tmp . > Server/User: http://fex.uni-stuttgart.de/frams...@rus.uni-stuttgart.de > /tmp/2GB.tmp : 2048 MB in 87 s (24105 kB/s) > > LXC: > > framstag@diaspora:~: fexsend -i flupp /tmp/2GB.tmp . > Server/User: http://flupp/frams...@rus.uni-stuttgart.de > /tmp/2GB.tmp : 2048 MB in 24 s (87381 kB/s) > > The ESX host has about twice of native CPU and disk power than the LXC > host. ESX has 10 Gb/s, LXC has 1 GB/s. Both VMs run the same software.
But that is to be expected, is it? There is no virtual i/O layer on lxc; ESX (or is it ESXi?) probably is not using SRIOV with that 10G NIC, etc. But lxc won't be able to run Windows (which is the main purpose of ESX(i)). Kind regards, Töns -- There is no safe distance. ------------------------------------------------------------------------------ What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Lxc-users mailing list Lxc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-users
