1 |
On Tue, Oct 18, 2011 at 12:53 AM, Florian Philipp <lists@×××××××××××.net> wrote: |
2 |
> Am 18.10.2011 07:16, schrieb Paul Hartman: |
3 |
>> On Wed, Oct 12, 2011 at 8:52 PM, Pandu Poluan <pandu@××××××.info> wrote: |
4 |
>>> Just stumbled upon this blog: |
5 |
>>> |
6 |
>>> http://www.webupd8.org/2011/10/increased-performance-in-linux-with.html |
7 |
>>> |
8 |
>>> anyone got any experience with zram/compcache on Gentoo? |
9 |
>> |
10 |
>> I'm using zram in a gentoo server with only 256mb of RAM, only used |
11 |
>> for a few weeks so far. It seems to work and the server hasn't crashed |
12 |
>> yet. :) I have allocated 128MB of compressed swap (64x2, actually, to |
13 |
>> theoretically utilize both CPU cores for compression at the same time) |
14 |
>> followed by normal on-disk swap at lower priority. Usually my total |
15 |
>> swap used is less than 128MB so the real disk swap is rarely touched. |
16 |
>> It's difficult to say if there is any improved performance, but I |
17 |
>> haven't experienced any slowdown, which occasionally I did when swap |
18 |
>> became heavily used in the past. Keep in mind the 128MB zram is the |
19 |
>> uncompressed size, so the actual amount of RAM used by this should be |
20 |
>> much less, depending on contents of the swap. Some even recommend |
21 |
>> using zram equal to the amount of RAM but that idea scares me. |
22 |
>> |
23 |
>> After enabling the CONFIG_ZRAM module in kernel 3.0.6, I did this: |
24 |
>> |
25 |
>> modprobe zram num_devices=2 |
26 |
>> echo $((64*1024*1024)) > /sys/block/zram0/disksize |
27 |
>> echo 1 > /sys/block/zram0/reset |
28 |
> # sleep 1 |
29 |
>> mkswap /dev/zram0 |
30 |
>> swapon -p 11 /dev/zram0 |
31 |
>> |
32 |
> |
33 |
> In my experience, it can be necessary to put a `sleep 1` between reset |
34 |
> and mkswap because the /dev/zram0 disappears and reappears after the |
35 |
> reset command. |
36 |
|
37 |
Good to know, thanks. In my case I typed those commands manually, so |
38 |
of course I didn't encounter any timing-related problem like that. |
39 |
|
40 |
> Another remark: The kernel docs recommend using /bin/echo instead of |
41 |
> echo because problems are reported as write errors and the echo builtin |
42 |
> of bash doesn't check for that. |
43 |
|
44 |
Also noted, thanks again. |