1 |
Yo Duncan! |
2 |
|
3 |
On Fri, 28 Jun 2013 09:12:24 +0000 (UTC) |
4 |
Duncan <1i5t5.duncan@×××.net> wrote: |
5 |
|
6 |
> Duncan posted on Fri, 28 Jun 2013 03:36:10 +0000 as excerpted: |
7 |
> |
8 |
> I settled on a 4 GiB file. Speeds are power-of-10-based since that's |
9 |
> what dd reports, unless otherwise stated. |
10 |
|
11 |
dd is pretty good at testing linear file performance, pretty useless |
12 |
for testing mysql performance. |
13 |
|
14 |
> To SSD: peak was upper 250s MB/s over a wide blocksize range of 1 MiB |
15 |
> >From SSD: peak was lower 480s MB/s, blocksize 32 KiB to 512 KiB |
16 |
|
17 |
Sounds about right. Your speeds are now so high that small differences |
18 |
in the SATA controller chip will be bigger than that between some |
19 |
SSD drives. Use a PCIe/SATA card and your performance will drop from |
20 |
what you see. |
21 |
|
22 |
> Spinning rust speeds, single Seagate st9500424as, 7200rpm 2.5" 16MB |
23 |
|
24 |
Those are pretty old and slow. If you are going to test an HDD against a |
25 |
newer SSD you should at least test a newer HDD. A new 2TB drive could |
26 |
get pretty close to your SSD performance in linear tests. |
27 |
|
28 |
> To rust: upper 70s MB/s, blocksize didn't seem to matter much. |
29 |
> >From rust: upper 90s MB/s, blocksize upto 4 MiB. |
30 |
|
31 |
Seems about right, for that drive. |
32 |
|
33 |
I think your numbers are about right, if your workload is just reading |
34 |
and writing big linear files. For a MySQL workload there would be a lot |
35 |
of random reads/writes/seeks and the SSD would really shine. |
36 |
|
37 |
RGDS |
38 |
GARY |
39 |
--------------------------------------------------------------------------- |
40 |
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701 |
41 |
gem@××××××.com Tel:+1(541)382-8588 |