Benchmarking virtual Network card on KVM

bukharyi · **Joined:** Tue Nov 03, 2009 11:29 pm **Posts:** 3

Hi,

I am doing benchmarking on virtualization technology targeting specifically on Linux KVM. CPU, Memory and disk i could benchmark, but for certain reason i got lost on how to benchmark virtual network card emulated by KVM.

Does anyone have an idea how to benchmark virtual network card? The main goal is, is the performance of virtual network card is comparable to physical machine network card?

ibarrere · **Posted:** Wed Nov 04, 2009 12:03 am

You could use something like iperf or lmbench. I would go with iperf, easy to use and install and quite powerful for a simple network throughput test.

bukharyi · **Joined:** Tue Nov 03, 2009 11:29 pm **Posts:** 3

Ahaa thank you. I use netperf which is simillar to the tools you stated. I can detect throughput and latency on the network.

The problem is, if i compare a 100Mbps virtual NIC with 100Mbps physical NIC, is it comparable?

mellowd · **Posted:** Wed Nov 04, 2009 3:44 am

bukharyi wrote:

The problem is, if i compare a 100Mbps virtual NIC with 100Mbps physical NIC, is it comparable?

If you test it you'll have the answer

bukharyi · **Joined:** Tue Nov 03, 2009 11:29 pm **Posts:** 3

I will and maybe i could post up the results.

alixen · **Joined:** Thu Nov 05, 2009 4:37 am **Posts:** 1

Hi,

I've done some IO disk benchmark with KVM, and i was a bit surprised by the results ...
Will you publish your result ?

Quote:

I'm currently testing some distribution with uptodate/new KVM :
- Ubuntu Karmic 9.10
- Rhel 5.4

I'm using IoZone to test IO disk performance.
For my test I have dedicated a LVM partition to the IoZone benchmark :
iozone -a -U /mnt/bench/ -f /mnt/bench/test-file -R -b <file.xls>

I have done multiple tests with 'cache=none', cache='writeback' and
cache='writethrough'.
Inside the VM I have tried, for the filesystem, the options data=ordered and
data=writeback.

All my tests have different results but at last, something is strange : in a
VM, IO Disk performance are better than in the host system for little files and
blocks and worse for big files and blocks.

I can give all my iozone benchmark results if you'd like

.

Well I do some assumption :
- KVM provide a cache for read/write operations
- KVM tells to the guest that data are written, but in fact, they are not.

Can someone explain me this behaviour ? Can I control it ?
Can this lead to data corruption in case of hardware crash ?

More informations
==============================
Ubuntu 9.10
-----------------------------------
- host : Ubuntu 9.10 ; kernel 2.6.31 ; qemu-kvm 0.11 ; lvm dedicated partition
in ext3 (data=ordered or writeback).

- guest : Ubuntu 9.10 kernel 2.6.31 ; qemu-kvm 0.11 ; virtio disk; lvm
dedicated partition in ext3 (data=ordered or writeback).

- The launch line :
/usr/bin/kvm -M pc-0.11 -m 512 -smp 1 -name kvm-ubuntu910 -uuid e5a362c5-
c28a-93dd-043b-d46eb4daba37 -monitor unix:/var/run/libvirt/qemu/kvm-
ubuntu910.monitor,server,nowait -boot c -drive file=/dev/storage-local-vol2-
lvm/kvm-ubuntu910,if=virtio,index=0,cache=<cache_option>,boot=on -drive
file=/dev/storage-local-vol2-lvm/bench,if=virtio,cache=<cache_option>,index=1 -
k fr

I have done tests with cache='none', cache='writeback' and
cache='writethrough'. Everytime, the results was différents but in the same way
: in the VM, IO Disk performance are better for little files and blocks and
worst for big files and blocks.

RHEL5.4
==============================
Similar tests ... Similar behaviour ...
Host : RHEL5.4, ext3, KVM from Redhat; lvm dedicated partitions; driver disk
from Redhat
Guest : RHEL5.4, ext3, KVM from Redhat; lvm dedicated partitions; driver disk
from Redhat

Regards,

Benchmarking virtual Network card on KVM

Who is online