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

[George Dunlap <george.dunlap@xxxxxxxxxx>]

On 25/11/15 09:27, Dario Faggioli wrote:
> [Actually Cc-ing George, using the right address, this time!]
> 
> -------- Forwarded Message --------
> Subject: Re: [Xen-users] Xen VMs and Unixbench: single vs multiple cpu
> behaviour
> Date: Tue, 24 Nov 2015 23:41:34 +0100
> Mailer: Evolution 3.18.2 (3.18.2-1.fc23) 
> Message-Id: <1448404894.7833.74.camel@xxxxxxxxxx>
> 
> [Cc-ing George, which I should have done earlier, sorry! :-/]
> 
> On Sat, 2015-11-21 at 11:41 +0100, Marko ÄukiÄ wrote:
>> And the results of vcpu pinning:
>>
> So, let me see if I can put the numbers together and recap.
> 
> With a 4 vCPUs VM, we have:
>                                     no pinning / all on 1 pCPU / 1-to-1 pin
> Dhrystone 2 using register variables  3355.0       3359.4          3385.2
> Double-Precision Whetstone             787.6        785.3           784.2
> Execl Throughput                       298.8        193.0           303.7
> File Copy 1024 bufsize 2000 maxblocks 3292.7       3303.1          3294.0
> File Copy 256 bufsize 500 maxblocks   2078.2       2089.2          2083.3
> File Copy 4096 bufsize 8000 maxblocks 5516.9       5559.8          5576.7
> Pipe Throughput                       1855.9       1857.8          1856.1
> Pipe-based Context Switching           999.9        987.6           999.5
> Process Creation                       254.4        826.4           354.1
> Shell Scripts (1 concurrent)           818.0        840.1           815.8
> Shell Scripts (8 concurrent)          6493.1       1100.4          6497.7
> System Call Overhead                  2870.2       2866.0          2847.9
> System Benchmarks Index Score         1564.2       1438.5          1611.2

Just to be clear -- for all of these numbers, higher is better, is that
correct?

 -George


_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users