Re: measuring RAID arrays performances

2013-05-28 Thread Stan Hoeppner
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 ? T

Re: measuring RAID arrays performances

2013-05-28 Thread Jean-Marc
On Sun, 26 May 2013 19:21:57 -0500, Stan Hoeppner wrote: Hi Stan, hi everybody, > Hopefully I'm teaching not only you, but others, that trying to > optimized performance of a -2 disk- array is a waste of -your- time. I have to say I was a little bit disappointed after this. I took time to read

Re: measuring RAID arrays performances

2013-05-26 Thread Stan Hoeppner
On 5/26/2013 5:40 AM, Jean-Marc wrote: > OK. Thank's for your mail and sorry to make you loosing your time reading > mine. Don't apologize. I'm not wasting my time here. Quite the opposite. Hopefully I'm teaching not only you, but others, that trying to optimized performance of a -2 disk- arr

Re: measuring RAID arrays performances

2013-05-26 Thread Stan Hoeppner
On 5/26/2013 4:26 AM, Richard Hector wrote: > On 26/05/13 15:15, Stan Hoeppner wrote: >> First there is no such thing as "RAID0". Second, "striping" is only fit >> for temporary data. The OP is concerned with permanent data storage. > >>From mdadm(8): > >> Currently, Linux supports LINEAR md de

Re: measuring RAID arrays performances

2013-05-26 Thread Andrei POPESCU
On Du, 26 mai 13, 00:36:43, Jean-Marc wrote: > Hi guys, > > I am trying to measure the perf' of RAID arrays to be sure of the > layout I'll use for my future set up. > > Based on this article, RAID10,f2 seems to be the best set up: > https://raid.wiki.kernel.org/index.php/Performance That articl

Re: measuring RAID arrays performances

2013-05-26 Thread Jean-Marc
On Sat, 25 May 2013 18:32:59 -0500 Stan Hoeppner wrote: Hi Stan, > > May be I am doing something wrong. > > Yes, you are. You're using only 2 disk drives. The md/RAID10 module > with two drives will perform nearly identical to RAID1. > > Adding insult to injury you are using 6GB partitions i

Re: measuring RAID arrays performances

2013-05-26 Thread Jean-Marc
On Sat, 25 May 2013 18:32:59 -0500 Stan Hoeppner wrote: Hi Stan, > > May be I am doing something wrong. > > Yes, you are. You're using only 2 disk drives. The md/RAID10 module > with two drives will perform nearly identical to RAID1. > > Adding insult to injury you are using 6GB partitions i

Re: measuring RAID arrays performances

2013-05-26 Thread Richard Hector
On 26/05/13 15:15, Stan Hoeppner wrote: > First there is no such thing as "RAID0". Second, "striping" is only fit > for temporary data. The OP is concerned with permanent data storage. >From mdadm(8): > Currently, Linux supports LINEAR md devices, RAID0 (striping), > RAID1 (mirroring), RAID

Re: measuring RAID arrays performances

2013-05-25 Thread Stan Hoeppner
On 5/25/2013 9:30 PM, green wrote: > Stan Hoeppner wrote at 2013-05-25 18:32 -0500: >> For two disks, use RAID1. PERIOD. End of story. Screwing with >> anything else is a massive waste of your time. If you have 3 or more >> disks, then you can worry about layouts. > > Or RAID0? That is, for 2

Re: measuring RAID arrays performances

2013-05-25 Thread green
Stan Hoeppner wrote at 2013-05-25 18:32 -0500: > For two disks, use RAID1. PERIOD. End of story. Screwing with > anything else is a massive waste of your time. If you have 3 or more > disks, then you can worry about layouts. Or RAID0? That is, for 2 disks use RAID0 (performance gain) *or* RAI

Re: measuring RAID arrays performances

2013-05-25 Thread Stan Hoeppner
On 5/25/2013 5:36 PM, Jean-Marc wrote: > Hi guys, > > I am trying to measure the perf' of RAID arrays to be sure of the layout > I'll use for my future set up. > > Based on this article, RAID10,f2 seems to be the best set up: > https://raid.wiki.kernel.org/index.php/Performance > > But if I crea