Re: measuring RAID arrays performances
On 5/28/2013 4:37 AM, Jean-Marc wrote:
...
> Back to what I read, I have a question:
> - are RAID0 and RAID10,f2 performances quiet similar in term of random
> read or not ?
Similar, yes. Identical, no.
> - and if there are not, why people are publish these kind of infos on
> the mdadm wiki ?
That's a question for the wiki author.
> - moreover, regarding the daily usage of a desktop, does a usefull RAID
> layout exist to give performances benefit ?
Sure. But you have only two disks and you desire redundancy. Thus
you're very limited in performance scalability. Which has been my point
throughout this thread. You are attempting to optimize the performance
of a very low performance system, thus any gains will be small.
> Advanced users ? I would say experienced users.
> Because even if you take your time to document yourself, there are few
> things helping you to take the right decision.
> But may be I was looking in the wrong direction.
I'd say the latter. There simply are not big gains to be had in
optimizing the RAID layout of -2- disks.
> Anyway, I am busy setting up some small VMs to test some services
> within isolated environments.
> I was thinking running them on my small partition using different RAID
> layouts to see how it behaves in real world.
>
> It will cost nothing extra and will help me to get ... more
> experienced.
> :-)
Yes, it will only cost your time. Most people's time is better spent on
things other than optimizing a 2 disk array. Maybe yours is not.
> Anyway, thank you for all the infos you gave through this exchange.
I was hoping to prevent you from wasting your time, but apparently I
have failed. If you want to use RAID10,f2 get yourself one more disk
drive, in which case it makes sense to use it.
--
Stan
Reply to: