1 |
Am Sonntag 31 Juli 2011, 19:11:06 schrieb Michael Mol: |
2 |
|
3 |
> > which is another goof reason not to use firefox - but |
4 |
> > total used free shared buffers cached |
5 |
> > Mem: 8182556 7373736 808820 0 56252 |
6 |
> > 2197064 |
7 |
> > -/+ buffers/cache: 5120420 3062136 |
8 |
> > Swap: 23446848 82868 23363980 |
9 |
> > |
10 |
> > even with lots of ram, you will hit swap. And since you are using the |
11 |
> > wear- leveling of the drive's firmware it does not matter that your |
12 |
> > swap resides on its own partition - every page written means a |
13 |
> > block-rewrite somewhere. Really not good for your ssd. |
14 |
> |
15 |
> Fair enough. |
16 |
> |
17 |
> It Would Be Nice(tm) if the SSD's block size and alignment matched |
18 |
> that of the kernel's pagesize. Not certain if it's possible to tune |
19 |
> those settings (reliably) in the kernel. |
20 |
> |
21 |
|
22 |
it would be enough if the write/delete blocks were a little bit smaller.. |
23 |
|
24 |
> Yet would potentially run afoul of the SSD's write block resolution. |
25 |
> And, of course, having the journal fail out from under me would be a |
26 |
> fair bit worse than the kernel panicking during a swap operation. |
27 |
> |
28 |
|
29 |
SSDs have this nice feature - it is called sudden violent death. Good chance |
30 |
that when it fails, all of it fails ;) |
31 |
|
32 |
I am backing up my ssd once a week. One week is the amount of emails and non- |
33 |
essential documents I am willing to risk. |
34 |
|
35 |
> >> Did you miss the last week's worth of discussion of memory limits on |
36 |
> >> tmpfs?> |
37 |
> > probably. Because I am using tempfs for /var/tmp/portage for ages and |
38 |
> > the only problematic packet is openoffice/libreoffice. |
39 |
> |
40 |
> I ran into trouble with Thunderbird a couple months ago, which is why |
41 |
> I had to drop from using tmpfs. (Also, I compile with -ggdb in CFLAGS, |
42 |
> so I expect my build sizes bloat a bit more than most) |
43 |
|
44 |
uuuh.. yeah.. 'a bit'.. you are a man of understatement. |
45 |
|
46 |
> |
47 |
> Anyway, the edge cases and caveats like the ones discussed are why I |
48 |
> ask about what people have tried, and what mitigators, workarounds and |
49 |
> technological improvements people have been working on. |
50 |
|
51 |
Well, you could always get a couple of 15k rpm u320 drives in a RAID 10 setup |
52 |
and use that for swap. |
53 |
|
54 |
Fucking fast (I have two in a spare box) but the noise is... extrem. |
55 |
|
56 |
|
57 |
-- |
58 |
#163933 |