[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] Xen VMs and Unixbench: single vs multiple cpu behaviour



Multiple CPU, huh?

Might be a NUMA issue.

On Sun, Nov 15, 2015 at 6:53 PM Marko ÄukiÄ <marko.djukic@xxxxxxxxx> wrote:
I am using a 64-bit host with a 64-bit guest (ubuntu server 14.04.2).

This are the contents of the .cfg file:

name = "ubuntu"
uuid = "4dea24b3-1d52-d8f3-2516-782e98a23fa0"
maxmem = 7168
memory = 7168
vcpus = 4
builder = "hvm"
boot = "c"
pae = 1
acpi = 1
apic = 1
hap = 1
viridian = 0
rtc_timeoffset = 0
localtime = 0
>
>
>
sdl = 0
vnc = 1
vncunused = 1
disk = [ "file:ubuntu_server_14.04_test.raw,xvda,w" ]
vif = [ "mac=24:42:53:21:52:45,bridge=br0,script=vif-bridge" ]
parallel = "none"
serial = "none"

I am new at Xen so maybe I missed a setting?Â

Regards

Marko

On 15 November 2015 at 08:37, Sarah Newman <srn@xxxxxxxxx> wrote:
On 11/14/2015 11:03 PM, Marko ÄukiÄ wrote:
> Hello!
> I am doing a comparison of different hypervisors using the Unixbench benchmark software.
> The testing is done on a machine with an Intel Core i5 cpu (4 cores).
>
> I am running two series of tests:
> - test A: 10 repetitions of the unixbench test in a VM with 1 cpu
> - test B: 10 repetitions of the same test in a VM with 4 cpus
>
> Test B has two sets of results because in a multicpu machine Unixbench runs a
>Â single test first (let's call it B-1) and then runs 4 tests in parallel - one for each cpu (test B-2).
>
> Compared to other hypervisors, in Xen VMs the results of tests A and B-2 are very good.
> Results of B-1, however, are much worse at tests with system calls.
> For example:
>
> Xen VM:
> test A create process test average score: 1871,7
> test B-1 create process test average score: 286,7 (!)
> test B-2 create process test average score:3259,2
> Qemu/KVM VM:
> test A create process test average score: 1865,6
> test B-1 create process test average score: 1095,3
> test B-2 create process test average score:3271,7
>
> Comparing total unixbench score Xen outperforms Qemu/KVM in tests A and B-2.
>
> Can anybody explain such a large difference in performance in test B-1 in Xen
> compared to Qemu/KVM?

I can guess, but to confirm is it 32 or 64 bit and what type of guest are you using? See
http://wiki.xen.org/wiki/Xen_Project_Software_Overview#Guest_Types

Slide 36 of http://events.linuxfoundation.org/sites/events/files/slides/pvh-linux-collab-summit_0.pdf (March 2014) compares PV, HVM, and PVH.

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users
--
Thanks,
Cheng Kun
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.