1 |
Hi, |
2 |
I got a WD 1T drive to use in a new machine for my dad. I didn't |
3 |
pay a huge amount of attention to the technical details when I |
4 |
purchased it other than it was SATA2, big, and the price was good. |
5 |
Here's the NewEgg link: |
6 |
|
7 |
http://www.newegg.com/Product/Product.aspx?Item=N82E16822136490 |
8 |
|
9 |
I installed the drive, created some partitions and set off to put |
10 |
ext3 on it using just mke2fs -j /dev/sda3. The partitions gets written |
11 |
and everything works but when I started installing Gentoo on it I was |
12 |
getting some HUGE delays at times, such as when unpacking |
13 |
portage.latest.tar.bz. Basically the tar step would be rolling along |
14 |
and then the drive would literally appear to stop for 1 minute before |
15 |
proceeding. No CPU usage, the machine is alive in other terminals, but |
16 |
anything directed at the disk just seems dead. Sticking my ear on the |
17 |
drive it doesn't sound like the drive is doing anything. |
18 |
|
19 |
I was trying to determine what to do - I.e is this a bad drive, how |
20 |
to return it, etc. - and started reading the reviews at NewEgg. One |
21 |
guy using it with Linux had this to say: |
22 |
|
23 |
<QUOTE> |
24 |
4KB physical sectors: KNOW WHAT YOU'RE DOING! |
25 |
|
26 |
Pros: Quiet, cool-running, big cache |
27 |
|
28 |
Cons: The 4KB physical sectors are a problem waiting to happen. If you |
29 |
misalign your partitions, disk performance can suffer. I ran |
30 |
benchmarks in Linux using a number of filesystems, and I found that |
31 |
with most filesystems, read performance and write performance with |
32 |
large files didn't suffer with misaligned partitions, but writes of |
33 |
many small files (unpacking a Linux kernel archive) could take several |
34 |
times as long with misaligned partitions as with aligned partitions. |
35 |
WD's advice about who needs to be concerned is overly simplistic, |
36 |
IMHO, and it's flat-out wrong for Linux, although it's probably |
37 |
accurate for 90% of buyers (those who run Windows or Mac OS and use |
38 |
their standard partitioning tools). If you're not part of that 90%, |
39 |
though, and if you don't fully understand this new technology and how |
40 |
to handle it, buy a drive with conventional 512-byte sectors! |
41 |
</QUOTE> |
42 |
|
43 |
Now, I don't mind getting a bit dirty learning to use this |
44 |
correctly but I'm wondering what that means in a practical sense. |
45 |
Reading the mke2fs man page the word 'sector' doesn't come up. It's my |
46 |
understanding the Linux 'blocks' are groups of sectors. True? If the |
47 |
disk must use 4K sectors then what - the smallest block has to be 4K |
48 |
and I'm using 1 sector per block? It seems that ext3 doesn't support |
49 |
anything larger than 4K? |
50 |
|
51 |
As a test I blew away all the partitions and made one huge 1 |
52 |
terabyte partition using ext3. I think tried untarring the portage |
53 |
snapshot and then deleting the directory where I put it a bunch of |
54 |
times. I get very different times each time I do this. untarring |
55 |
varies from 6 minutes 24 seconds to 10 minutes 25 seconds. Removing |
56 |
the directory varies from 3 seconds to 1 minute 22 seconds. |
57 |
|
58 |
Every time there is an apparent delay I just see the hard drive |
59 |
light turned on solid. That said as far as I know if I wait for things |
60 |
to complete the data is there but I haven't tested it extensively. |
61 |
|
62 |
Is this a bad drive or am I somehow using it incorrectly? |
63 |
|
64 |
Thanks, |
65 |
Mark |
66 |
|
67 |
|
68 |
gandalf TestMount # time tar xjf /mnt/TestMount/portage-latest.tar.bz2 |
69 |
-C /mnt/TestMount/usr |
70 |
|
71 |
real 6m24.736s |
72 |
user 0m9.969s |
73 |
sys 0m3.537s |
74 |
gandalf TestMount # time rm -rf /mnt/TestMount/usr/ |
75 |
|
76 |
real 0m3.229s |
77 |
user 0m0.110s |
78 |
sys 0m1.809s |
79 |
gandalf TestMount # mkdir usr |
80 |
gandalf TestMount # time tar xjf /mnt/TestMount/portage-latest.tar.bz2 |
81 |
-C /mnt/TestMount/usr |
82 |
|
83 |
real 7m50.193s |
84 |
user 0m8.647s |
85 |
sys 0m2.811s |
86 |
gandalf TestMount # time rm -rf /mnt/TestMount/usr/ |
87 |
|
88 |
real 0m3.234s |
89 |
user 0m0.119s |
90 |
sys 0m1.792s |
91 |
gandalf TestMount # mkdir usr |
92 |
gandalf TestMount # time tar xjf /mnt/TestMount/portage-latest.tar.bz2 |
93 |
-C /mnt/TestMount/usr |
94 |
|
95 |
real 10m25.926s |
96 |
user 0m8.645s |
97 |
sys 0m2.765s |
98 |
gandalf TestMount # time rm -rf /mnt/TestMount/usr/ |
99 |
|
100 |
real 1m22.330s |
101 |
user 0m0.124s |
102 |
sys 0m1.810s |
103 |
gandalf TestMount # mkdir usr |
104 |
gandalf TestMount # time tar xjf /mnt/TestMount/portage-latest.tar.bz2 |
105 |
-C /mnt/TestMount/usr |
106 |
|
107 |
real 8m12.307s |
108 |
user 0m8.463s |
109 |
sys 0m2.708s |
110 |
gandalf TestMount # time rm -rf /mnt/TestMount/usr/ |
111 |
|
112 |
real 0m29.517s |
113 |
user 0m0.114s |
114 |
sys 0m1.810s |
115 |
gandalf TestMount # |
116 |
|
117 |
|
118 |
|
119 |
|
120 |
gandalf ~ # hdparm -tT /dev/sdb |
121 |
|
122 |
/dev/sdb: |
123 |
Timing cached reads: 11362 MB in 2.00 seconds = 5684.46 MB/sec |
124 |
Timing buffered disk reads: 314 MB in 3.00 seconds = 104.64 MB/sec |
125 |
gandalf ~ # |