1 |
On Sun, Feb 7, 2010 at 9:30 AM, Alexander <b3nder@××××××.ru> wrote: |
2 |
> On Sunday 07 February 2010 19:27:46 Mark Knecht wrote: |
3 |
> |
4 |
>> Every time there is an apparent delay I just see the hard drive |
5 |
>> light turned on solid. That said as far as I know if I wait for things |
6 |
>> to complete the data is there but I haven't tested it extensively. |
7 |
>> |
8 |
>> Is this a bad drive or am I somehow using it incorrectly? |
9 |
>> |
10 |
> |
11 |
> Is there any related info in dmesg? |
12 |
> |
13 |
> |
14 |
|
15 |
No, nothing in dmesg at all. |
16 |
|
17 |
Here are two tests this morning. The first is to the 1T drive, the |
18 |
second is to a 120GB drive I'm currently using as a system drive until |
19 |
I work this out: |
20 |
|
21 |
gandalf TestMount # time tar xjf /mnt/TestMount/portage-latest.tar.bz2 |
22 |
-C /mnt/TestMount/usr |
23 |
|
24 |
real 8m13.077s |
25 |
user 0m8.184s |
26 |
sys 0m2.561s |
27 |
gandalf TestMount # |
28 |
|
29 |
|
30 |
mark@gandalf ~ $ time tar xjf /mnt/TestMount/portage-latest.tar.bz2 -C |
31 |
/home/mark/Test_usr/ |
32 |
|
33 |
real 0m39.213s |
34 |
user 0m8.243s |
35 |
sys 0m2.135s |
36 |
mark@gandalf ~ $ |
37 |
|
38 |
8 minutes vs 39 seconds! |
39 |
|
40 |
The amount of data written appears to be the same: |
41 |
|
42 |
gandalf ~ # du -shc /mnt/TestMount/usr/ |
43 |
583M /mnt/TestMount/usr/ |
44 |
583M total |
45 |
gandalf ~ # |
46 |
|
47 |
|
48 |
mark@gandalf ~ $ du -shc /home/mark/Test_usr/ |
49 |
583M /home/mark/Test_usr/ |
50 |
583M total |
51 |
mark@gandalf ~ $ |
52 |
|
53 |
|
54 |
I did some reading at the WD site and it seems this drive does use the |
55 |
4K sector size. The way it's done is the addressing on cable is still |
56 |
512 byte 'user sectors', but they are packed into 4K physical sectors |
57 |
and internal hardware does the mapping. |
58 |
|
59 |
I suspect the performance issue is figuring out how to get the file |
60 |
system to keep things on 4K boundaries. I assume that's what the 4K |
61 |
block size is for when building the file system but I need to go find |
62 |
out more about that. I did not select it specifically. Maybe I need |
63 |
to. |
64 |
|
65 |
Thanks, |
66 |
Mark |