1 |
On 8 Feb 2010, at 05:25, Valmor de Almeida wrote: |
2 |
|
3 |
> Mark Knecht wrote: |
4 |
>> On Sun, Feb 7, 2010 at 11:39 AM, Willie Wong <wwong@××××××××××××××.edu |
5 |
>> > wrote: |
6 |
> [snip] |
7 |
>> OK - it turns out if I start fdisk using the -u option it show me |
8 |
>> sector numbers. Looking at the original partition put on just using |
9 |
>> default values it had the starting sector was 63 - probably about the |
10 |
> |
11 |
> I too was wondering why a Toshiba HDD 1.8" MK2431GAH (4kB-sector), 240 |
12 |
> GB I've recently obtained was slow: |
13 |
> |
14 |
> -> time tar xfj portage-latest.tar.bz2 |
15 |
> |
16 |
> real 16m5.500s |
17 |
> user 0m28.535s |
18 |
> sys 0m19.785s |
19 |
> |
20 |
> Following your post I recreated a single partition (reiserfs 3.6) |
21 |
> starting at the 64th sector: |
22 |
> |
23 |
> Disk /dev/sdb: 240.1 GB, 240057409536 bytes |
24 |
> 255 heads, 63 sectors/track, 29185 cylinders, total 468862128 sectors |
25 |
> Units = sectors of 1 * 512 = 512 bytes |
26 |
> Disk identifier: 0xe7bf4b8e |
27 |
> |
28 |
> Device Boot Start End Blocks Id System |
29 |
> /dev/sdb1 64 468862127 234431032 83 Linux |
30 |
> |
31 |
> and the time was improved |
32 |
> |
33 |
> -> time tar xfj portage-latest.tar.bz2 |
34 |
> |
35 |
> real 2m15.600s |
36 |
> user 0m28.156s |
37 |
> sys 0m18.933s |
38 |
|
39 |
Thanks to both you & Mark for posting this information about these |
40 |
improved timings. |
41 |
|
42 |
I have just checked, and I am getting 3.5 - 6 minutes (real) to untar |
43 |
portage. I had blamed performance of this array on the fact that the |
44 |
RAID controller is an older model PCI card I got cheap(ish) off eBay, |
45 |
but I see it is also aligned beginning at sector 63. |
46 |
|
47 |
I'm not quite sure if this is cause of poor performance here, as the |
48 |
drives in this array are not quite as modern as yours - I'm guessing |
49 |
that at least a couple of the drives have been bought in the last 6 |
50 |
months, but they are only 500GB drives. However I guess it would only |
51 |
require one drive in the array to have 4K sectors and it would cause |
52 |
this kind of slowdown. I will try checking their spec now. |
53 |
|
54 |
This is the same server that caused me to post in relation to slow |
55 |
Samba transfers 3 weeks ago ("How to determine if a NIC is playing |
56 |
gigabit?"). I have still not yet tested thoroughly - there are always |
57 |
chores getting in the way! - but it seems like I was able to transfer |
58 |
the same files in about a third (or maybe even a quarter) the time at |
59 |
100mbit, between my laptop & desktop Macs. |
60 |
|
61 |
I am not immediately able to alter the partition layout, as I have |
62 |
scads of data on this array. In order to test I think I will need to |
63 |
create a second array, aligned optimally, and copy the data across. |
64 |
|
65 |
I had been recently thinking that 2TB drives are now 40% cheaper per |
66 |
gig than 500GB ones, so perhaps I will have to splash out on 3 of |
67 |
them. This seems rather a lot of money, but I could probably use the |
68 |
space. Hmmmn... actually 1TB are nearly as cheap as per gig - |
69 |
considering the eBaying of my current drives, those would make a lot |
70 |
of sense. |
71 |
|
72 |
Stroller. |
73 |
|
74 |
|
75 |
|
76 |
$ time tar xfj portage-latest.tar.bz2 |
77 |
|
78 |
real 6m3.128s |
79 |
user 0m37.810s |
80 |
sys 0m39.614s |
81 |
$ echo p | sudo fdisk -u /dev/sdb |
82 |
|
83 |
The number of cylinders for this disk is set to 182360. |
84 |
There is nothing wrong with that, but this is larger than 1024, |
85 |
and could in certain setups cause problems with: |
86 |
1) software that runs at boot time (e.g., old versions of LILO) |
87 |
2) booting and partitioning software from other OSs |
88 |
(e.g., DOS FDISK, OS/2 FDISK) |
89 |
|
90 |
Command (m for help): |
91 |
Disk /dev/sdb: 1500.0 GB, 1499968045056 bytes |
92 |
255 heads, 63 sectors/track, 182360 cylinders, total 2929625088 sectors |
93 |
Units = sectors of 1 * 512 = 512 bytes |
94 |
Disk identifier: 0x27a827a7 |
95 |
|
96 |
Device Boot Start End Blocks Id System |
97 |
/dev/sdb1 63 2929613399 1464806668+ 83 Linux |
98 |
|
99 |
Command (m for help): Command (m for help): Command (m for help): |
100 |
got EOF thrice - exiting.. |
101 |
$ |