1 |
the only thing that i've splitted is: |
2 |
/boot on a 100Mb partition, and this thing has saved me a lot of pain when |
3 |
something went wrong with the reiserfschecks when the pc ran out of energy |
4 |
with ext2 unjournaled, |
5 |
the /home partition, so that it could be used with different systems without |
6 |
reconfiguring with reiserfs, |
7 |
and a large file partition with xfs which has a very great large file |
8 |
usage.... |
9 |
i personally don't really think that splitting other / subfolders may have a |
10 |
great use on everyday use.... |
11 |
|
12 |
2007/7/20, Bob Sanders <rsanders@×××.com>: |
13 |
> |
14 |
> Bernhard Auzinger, mused, then expounded: |
15 |
> > Hi, |
16 |
> > |
17 |
> > as I have four hdd's in my computer, I was wondering if it does make |
18 |
> sense to |
19 |
> > source out some partitions/directories to a second hdd. |
20 |
> > |
21 |
> |
22 |
> There is no simple answer. It really depends upon a lot of factors - |
23 |
> controllers, |
24 |
> drives, file system, memory, system bus... |
25 |
> |
26 |
> > At the moment I have separate partitions for /var, /tmp and /usr/portage |
27 |
> (I |
28 |
> > feel portage is a lot faster since I've done this) on the same hdd. |
29 |
> > |
30 |
> > My question is if it makes sence to move these partitions to another |
31 |
> harddisk? |
32 |
> > |
33 |
> |
34 |
> Spreading them across drives could result in faster access if the |
35 |
> controllers |
36 |
> the drives are atached to allow overlapping commands. IDE doesn't do this |
37 |
> and |
38 |
> can only have one drive active on the bus. |
39 |
> |
40 |
> Also, some things - /var/tmp/portage, need fast read/write access while |
41 |
> /usr/portage |
42 |
> is a large number of small files (things like Open Office, gcc, firefox |
43 |
> being |
44 |
> exceptions) and mainly read only access. In many cases /tmp is mainly an |
45 |
> initial |
46 |
> write, then mostly read only access once the file get built. |
47 |
> |
48 |
> Also, it depends on where on the drive the partition resides. Swap is |
49 |
> usually best |
50 |
> around one third to one half into the drive if the drive is 60% or more |
51 |
> full. Less |
52 |
> access time as the head idles arounf the swap area. |
53 |
> |
54 |
> And as drives fill up, they will slow down as file seek times |
55 |
> increase. Additionally, |
56 |
> different file systems will respond to the types of files differently - |
57 |
> lots of |
58 |
> small files, large streaming media files, indexed databases, all require |
59 |
> considering |
60 |
> the intended use. |
61 |
> |
62 |
> In desktop use, I've watched the typical file i/o and can say it tended to |
63 |
> stay |
64 |
> below 4 or 5 MB/s peak for most things. And I've seen raid 5 rebuilds |
65 |
> sustain |
66 |
> 225 MB/s on the same system. |
67 |
> |
68 |
> So, sure move the r/w tasks off disks competing with other r/w tasks and |
69 |
> leave the |
70 |
> read only tasks on the man system disk. You'll see a small improvement, |
71 |
> but in the |
72 |
> larger scheme of things, outside of uncompressing the kernel, open office, |
73 |
> firefox, |
74 |
> or gcc, it won't matter much more than 1% or 2% improvement. |
75 |
> |
76 |
> Bob |
77 |
> - |
78 |
> -- |
79 |
> gentoo-amd64@g.o mailing list |
80 |
> |
81 |
> |
82 |
|
83 |
|
84 |
-- |
85 |
beso |
86 |
|
87 |
d-_-b |