New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Why not let @serverian do it - he is knowledgable in multiple virts and not a biased Xen fanboy
@c0y I think your missing half a conversation here
He is going to do the KVM setup, I am going to do the Xen setup, the actual benchmarks will be ran by someone independent of both of us to ensure it was non bias, we are both pretty excited about giving the facts to the community regardless of outcome and final numbers
Ooh, my apologies, that is a great idea! :-)
I have a spare dedi that I can supply. It's @ Delimiter w/ 16GB RAM, 500GB HDD, iLo.
No idea how well it would work with VT.
Not to derail the topic at hand, but does turning off NCQ give that much of a performance improvement? I always thought having it enabled would help.
I don't honestly know, that's why I'm asking...
$10 on XEN, don't let me down
I don't care about the outcome and final numbers. All I care is that both of you are among the best providers that are about to show a demonstration of how good you are in tweaking different types of virtualization
XEN 4.4 is going to be released in 17 days, I would suggest benchmarking Vmware or KVM first & then for XEN when 4.4 is released
Yeah, that would be really awesome! Thanks!!
Even though I agree with you on that both are great guys, I personally think the results do matter. It's basically the most interesting of all! We know they are capable, now it's time to compare the technologies. Some benchmarks and a real comparison would really be nice to post and share with the community.
Ok, I agree with this.
What I really mean with "I don't care about the outcome and final numbers" is actually in terms of winning or losing.
Both guys are great, and the technologies in comparison has their pluses and minuses.
IMHO, for commercial purpose it depends on the skill and experience of the person(s) who push each virtualization technology to the limit to enable users/customers get the best experience in using their service.
Yeah I will probably stick to 4.2/3 though as using PVH on 4.4 is a little unfair
I could go 4.4 and not use PVH though I suppose.
We shall see.
It helps if your using standard SATA disks yes, think of it as something used to optimise the timing of the head movement on the drive to be more efficient, in a raid setup, especially mdadm it is largely redundant though and can actually make things less optimal, there are lots of opinions on it to be honest but the truth is it is probably so small of a difference its not worth the hassle.
I mention it only to make sure the 2 setups are as similar as possible.
I'm selfishly hoping that @AnthonySmith is leaving the door open there to the possibility of benchmarking both versions (his valuable time permitting)
It would make more sense to bench 4.x and 3.x solusvm by default uses 3.x unless you do a manual install of Xen first, that is largely the reason it got a reputation as not being as fast.. they bundle an 8 year old hypervisor!