1 |
Yo Duncan! |
2 |
|
3 |
On Fri, 28 Jun 2013 00:20:45 +0000 (UTC) |
4 |
Duncan <1i5t5.duncan@×××.net> wrote: |
5 |
|
6 |
> > I have around 30 RAID1 sets in production right now. Some of them |
7 |
> > doing mostly reads and some mostly writes. Some are HDD and some |
8 |
> > SSD. The RAID sets are pushed pretty hard 24x7 and we watch the |
9 |
> > performance pretty closely to plan updates. I have collectd |
10 |
> > performance graphs going way back. |
11 |
> |
12 |
> So you're basically confirming it with normal usage as well, but have |
13 |
> documented performance history going pretty well all the way back. |
14 |
> Not the simple test script I was hoping for, but pretty impressive, |
15 |
> none-the- less. |
16 |
|
17 |
I find that 'hdparm -tT', 'dd' and 'bonnie++' will match up pretty well |
18 |
with what I see in production. Just be sure to use really large test |
19 |
file sizes with bonnie++ and dd. dd also needs a pretty large block size |
20 |
(bs=) and pretty large/fast source of bits when writing. |
21 |
|
22 |
With bonnie++ you can easily see the speed differences between raw disks |
23 |
and various RAID types. |
24 |
|
25 |
|
26 |
RGDS |
27 |
GARY |
28 |
--------------------------------------------------------------------------- |
29 |
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701 |
30 |
gem@××××××.com Tel:+1(541)382-8588 |