1 |
On Friday 19 November 2010 19:19:34 David W Noon wrote: |
2 |
> On Fri, 19 Nov 2010 17:00:04 +0100, Peter Humphrey wrote about Re: |
3 |
> |
4 |
> [gentoo-user] migrating disks (from mounts to disklabels: |
5 |
> >On Friday 19 November 2010 14:42:23 Neil Bothwick wrote: |
6 |
> >> On Fri, 19 Nov 2010 10:52:50 +0000, Mick wrote: |
7 |
> >> > Also primary partitions which he does not seem to be using at all |
8 |
> >> > have a slight edge over logical. |
9 |
> >> |
10 |
> >> Do you have any data on this? I generally use all logical partitions |
11 |
> >> but could be persuaded to rethink. |
12 |
> > |
13 |
> >Well there must be one level of indirection on first access, since the |
14 |
> >start of the logical partition has to be looked up in a "primary" |
15 |
> >partition, but I can't imagine that being needed more than once per |
16 |
> >reboot. |
17 |
> |
18 |
> Correct. |
19 |
> |
20 |
> The same applies to LVM2 or EVMS logical volumes: a small "lookup" |
21 |
> penalty (a few milliseconds) when the filesystem is first |
22 |
> activated/mounted, and as fast as the drive itself thereafter. |
23 |
|
24 |
Short of measuring the latency with some system (which I wouldn't know how) I |
25 |
have experimented with setting the /boot partition on primary and logical |
26 |
partitions and the difference (on a stopwatch) was measurable in seconds |
27 |
betweeen having said partition on a primary and having it on a logical. |
28 |
Furthermore, sda7 was slower than sda5. |
29 |
|
30 |
I haven't measured latencies for first mount and subsequent look ups. I |
31 |
thought that it would be the same every time a partition fs is being accessed, |
32 |
no? |
33 |
-- |
34 |
Regards, |
35 |
Mick |