1 |
On Sun, Feb 7, 2010 at 6:08 PM, Willie Wong <wwong@××××××××××××××.edu> wrote: |
2 |
> On Sun, Feb 07, 2010 at 01:42:18PM -0800, Mark Knecht wrote: |
3 |
>> OK - it turns out if I start fdisk using the -u option it show me |
4 |
>> sector numbers. Looking at the original partition put on just using |
5 |
>> default values it had the starting sector was 63 - probably about the |
6 |
>> worst value it could be. As a test I blew away that partition and |
7 |
>> created a new one starting at 64 instead and the untar results are |
8 |
>> vastly improved - down to roughly 20 seconds from 8-10 minutes. That's |
9 |
>> roughly twice as fast as the old 120GB SATA2 drive I was using to test |
10 |
>> the system out while I debugged this issue. |
11 |
> |
12 |
> That's good to hear. |
13 |
> |
14 |
>> I'm still a little fuzzy about what happens to the extra sectors at |
15 |
>> the end of a track. Are they used and I pay for a little bit of |
16 |
>> overhead reading data off of them or are they ignored and I lose |
17 |
>> capacity? I think it must be the former as my partition isn't all that |
18 |
>> much less than 1TB. |
19 |
> |
20 |
> As far as I know, you shouldn't worry about it. The |
21 |
> head/track/cylinder addressing is a relic of an older day. Almost all |
22 |
> modern drives should be accessed via LBA. If interested, take a look |
23 |
> at the wikipedia entry on Cylinder-Head-Sector and Logical Block |
24 |
> Addressing. |
25 |
> |
26 |
> Basically, you are not losing anything. |
27 |
> |
28 |
> Cheers, |
29 |
> |
30 |
> W |
31 |
> -- |
32 |
> Willie W. Wong wwong@××××××××××××××.edu |
33 |
> Data aequatione quotcunque fluentes quantitae involvente fluxiones invenire |
34 |
> et vice versa ~~~ I. Newton |
35 |
> |
36 |
> |
37 |
|
38 |
Hi, |
39 |
Yeah, a little more study and thinking confirms this. The sectors |
40 |
are 4K. WD put them on there. The sectors are 4K. |
41 |
|
42 |
Just because there might be extra physical space at the end of a |
43 |
track doesn't mean I can ever use it. |
44 |
|
45 |
The sectors are 4K and WD put them on there and they've taken ALL |
46 |
that into account already. They are 4K physically with ECC but |
47 |
accessible by CHS and by LBA in 512B chunks. The trick for speed at |
48 |
the OS/driver level is to make sure we are always grabbing 4K logical |
49 |
blocks from a single 4K physical sector off the drive. If we do it's |
50 |
fast. If we don't and start asking for a 4K block that isn't in a |
51 |
single 4K physical block then it becomes very slow as the drive |
52 |
hardware/firmware/processor has to do multiple reads and piece it |
53 |
together for us which is slow. (VERY slow...) By using partitions |
54 |
mapped to sector number values divisible by 8 we do this. (8 * 512B = |
55 |
4K) |
56 |
|
57 |
The extra space at the end of a track/cylinder is 'lost' but it was |
58 |
lost before we bought the drive because the sectors are 4K so there is |
59 |
nothing 'lost' by the choices we make in fdisk. I must remember to use |
60 |
fdisk -u to see the sector numbers when making the partitions and |
61 |
remember to do some test writes to the partition to ensure it's right |
62 |
and the speed is good before doing any real work. |
63 |
|
64 |
This has been helpful for me. I'm glad Valmor is getting better |
65 |
results also. |
66 |
|
67 |
I wish I had checked the title before I sent the original email it |
68 |
was supposed to be |
69 |
|
70 |
1-Terabyte drives - 4K sector sizes? -> bad performance so far |
71 |
|
72 |
Maybe sticking that here will help others when they Google for this later. |
73 |
|
74 |
Cheers, |
75 |
Mark |