some interesting observations on the relative performance of kvm vs. bhyve

[Aryeh Friedman]

On Sat, Feb 1, 2014 at 10:16 PM, Peter Grehan <grehan at freebsd.org> wrote:

> > I have 1 host that dual boots FreeBSD and Ubuntu 12.04.3 LTS and bhtyve
>
>> seems to be atleast 3 or 4 times faster with disk I/O then kvm using the
>> most stripped down command lines I can come up with.
>>
>
>  I'm guessing that the default cache mode for qemu in that release is
> "none". You may want to switch it to "writeback", which is what bhyve does
> by default (it can be changed with AHCI, see bhyve(8)).
>

Does this bring up the same power failure scenario issues mentioned in the
link you provided?    It seems like the only way to get reasonable
performance is to be essentially unsafe in guest writes to the host disk?
A question does the ability of FreeBSD to be able to better handle power
failure in general better then linux (it seems like every time there is a
unscheduled reboot on linux it messes up)?   This seems to be at odds with
my personal observations of bhyve via petitecloud which I routinely very
abruptly start/stop (petitecloud's "stop" is nothing more then killing the
hyperv and any cleanup needed) and except for the occasional need for a
fsck have not had an issue.   But it does not seem to be at odds with
OpenStack's experience
http://docs.openstack.org/admin-guide-cloud/content/ch_introduction-to-openstack-compute.html#section_nova-disaster-recovery-process


>  Lots of info on the web about Qemu block i/o cache modes e.g.
>
>
> http://pic.dhe.ibm.com/infocenter/lnxinfo/v3r0m0/index.jsp?topic=%2Fliaat%
> 2Fliaatbpkvmguestcache.htm
>
> later,
>
> Peter.
>



-- 
Aryeh M. Friedman, Lead Developer, http://www.PetiteCloud.org